Green Et Al-2018-Journal of Child Psychology and Psychiatry

PFI_12mmX178mm.
pdf + eps format
Journal of Child Psychology and Psychiatry 59:4 (2018), pp 424–443 doi:10.1111/jcpp.12892
Annual Research Review: The state of autism

intervention science: progress, target psychological
and biological mechanisms and future prospects
Jonathan Green,1,2,3,4 and Shruti Garg1,2,3,4
1
Division of Neuroscience and Experimental Psychology, School of Biological Sciences, Faculty of Biology, Medicine
and Health, University of Manchester, Manchester; 2Manchester Academic Health Science Centre, Manchester;
3
Manchester University NHS Foundation Trust, Manchester; 4Greater Manchester Mental Health NHS Foundation
Trust, Manchester, UK
Background: There has been recent systematic review of key evidence in psychosocial intervention in autism but
little review of biological treatments. Methods: We analyse the current literature from the perspective of intervention
and mechanism targets across social and biological development. Results: The overall quality of trials evidence in
autism intervention remains relatively low, despite some recent progress. Many treatments in common use have little
or no evidence base. This is very concerning in such an important disorder. A variety of psychosocial interventions
can show effect to improve some short-term effects on children’s immediate dyadic social interactions, for instance
with caregivers. But showing true effectiveness in this developmental disorder requires generalisation of such effects
into wider social contexts, on autism symptoms and in long-term progress in development. Only a few interventions
so far have begun to show this. A number of early phase interventions on biological targets have shown real promise,
but none has yet progressed to larger scale effectiveness trials on behavioural or symptom outcomes. Conclusions:
There has been enough progress in psychosocial intervention research now to be able to begin to identify some
evidence-based practice in autism treatment. To consolidate and improve outcomes, the next phase of intervention
research needs improved trial design, and an iterative approach building on success. It may also include the testing
of potential synergies between promising biological and psychosocial interventions. Keywords: Autism spectrum
disorders; intervention; neurobiology; parent-child interaction; parent training.
mechanism level is timely. It points the way forward

Introduction
to a new phase of autism intervention science
A recent systematic review by French and Kennedy
addressing potential synergies between psychologi-
(2018) gives a methodologically rigorous summary of
cal and biological intervention strategies in this
the key evidence for psychosocial interventions in
neurodevelopmental disorder.
autism. The purpose of this current research review
is complimentary. First, we provide analysis and
commentary on what the evidence such as reviewed
by French and Kennedy (2018) represents for the Progress in psychosocial intervention research
current state of autism intervention science as a Trial methods
whole, in terms of methodological developments,
An immediately striking feature of the contempora-
implicit or explicit intervention targets and common
neous systematic review by French and Kennedy
factors in treatment. We focus particularly on inter-
(2018) is how few of the studies they identified met
vention and mechanism targets in order to provide a
basic Cochrane quality criteria for absence of risk of
clarifying and unifying thread with which to compare
bias (in just six trials out of the 48 that met their
evidence across different interventions, to establish
inclusion criteria, even though most had been con-
precisely what overall progress has been made, and
ducted in the last few years). Another is how chaotic
what has been learnt about autism as a develop-
the autism intervention field is internationally, with
mental condition from intervention research. We aim
those 48 trials testing no less than 34 different
for this to point a way forward systematically to new
models of intervention, usually in underpowered
and refined mechanism targets in future.
small N trials from single sites.
Next, we extend the focus on mechanism to bridge
While there has been substantive progress in
between intervention work targeting psychosocial
recent years, it remains striking and paradoxical
development and emerging work on complimentary
that the intervention science in such an important
neural system targets. These two domains are usu-
developmental disorder continues to be so frag-
ally considered separately; it is the thesis of this
mented and methodologically fragile. This cannot
review that considering them together now at a
just be put down to lack of funding. The majority of
interventions tested have been behavioural or psy-
chosocial, areas in which there has been debate
Conflict of interest statement: No conflicts declared.
© 2018 The Authors. Journal of Child Psychology and Psychiatry published by John Wiley & Sons Ltd on behalf of Association for Child and
Adolescent Mental Health.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited.
doi:10.1111/jcpp.12892 The state of autism intervention science 425
(historically but also currently) as to the appropri- of multiple measured outcomes that has often been
ateness or viability of conducting rigorous trials. Any the norm. At a stroke this discipline, although
doubts as to the value and importance of Ran- arduous initially, would transform the validity and
domised Controlled Trials (RCT) methodology in this salience of the trial reporting.
area however should have been laid to rest. It turns A third key step will be the explicit investigation of
out that they are not only feasible but can have great mechanism, which will allow an iterative evolution
advantages in clarifying and adjusting for the many towards more refined treatments (Green, 2015;
potential confounds involved in properly testing Kraemer, Wilson, Fairburn, & Agras, 2002). The first
complex interventions within an equally complex part of this review consequently builds on French
phenotype (Dunn et al., 2015). It has been a genuine and Kennedy (2018) by focusing on the (explicit or
achievement for the field (in parallel certainly with implicit) mechanisms and treatment targets within
other areas of psychosocial treatment research) to the variety of treatments tested to date – particularly
have demonstrated the value of RCT methods and, in those studies reviewed by French and Kennedy
increasingly, the implementation of CONSORT (2018) as having relatively low risk of bias. We aim
reporting standards; but the French and Kennedy through this to provide a unifying view of the current
(2018) review shows how patchy this has been in research along with an agenda for moving forward
practice. Enormous sums of public and private with new and refined mechanism targets. We then
money are spent in health care systems across the point the way forward to some of these new potential
world on interventions that have no rigorous evi- targets in brain science in the second part of the
dence base. For instance, probably the most ubiqui- review.
tous interventions globally (under the umbrella of
Early Intensive Behavioural Intervention (EIBI) or
Co-construction of outcome targets
ABA) continue to be actively promoted on the basis of
single case designs alone, internationally considered A further achievement for autism intervention has
to be low levels of evidence in healthcare; a Cochrane been an emerging dialogue with user groups, advo-
review of EIBI (Reichow, Barton, Boyd, & Hume, cates and families towards coconstructing relevant
2012) found only one historical, small and equivocal outcome variables and outcome targets after inter-
RCT and the situation has not changed subse- vention. Autism has benefitted from active debates
quently. Consequently, such interventions and and dialogue between these groups, which should be
others do not feature at all in the French and creative for future progress. At the extreme some
Kennedy (2018) review or in relevant guidance such advocates object to any idea of intervention at all as
as UK NICE (National Institute for Health & Care impacting negatively on ‘autism identity’, wanting
Excellence, 2013). This is a truly troubling situation instead a focus on acceptance, social rights and
given the high profile and importance of autism in associated mental health. These latter objectives are
public health, and its economic and social costs. On of undeniable importance, but the review below
just one metric, the life span economic cost of autism addresses how interventions, theoretically targeted
per individual has been estimated as up to $2.4 at core early developmental processes, can impact
million in the United States and £1.5 million in the positively on key autism features and adaptation, even
United Kingdom (Buescher, Cidav, Knapp, & Man- in this highly heritable neurodevelopmental disorder.
dell, 2014; Knapp, Romeo, & Beecham, 2009), This empirical evidence, alongside early diagnosis,
higher than asthma or diabetes. The reader might forms the ethical case for intervention; but the out-
imagine if international clinical practice in these comes selected do need to be of shared relevance. This
latter conditions was underpinned in this way. is a theme we return to at the end of the review.
In a further paradox, the autism field actually has
a significant advantage over some other areas of
psychosocial intervention research in having devel-
Targets and mechanisms in psychosocial
oped relatively valid, predictive and blind-codable
autism intervention
measures of social interaction and autism symptom
Parent-child dyadic interaction as a foundation for
outcomes; something that has often been elusive in
social development
other areas. Lack of assessment blinding in self- or
informant- reporting can seriously inflate treatment The rationale for targeting early parent-child social
effect estimates, especially in trials without partici- interaction (PCI) in autism as a social impairment
pant blinding to treatment, inevitably the norm in disorder comes from decades of child development
most psychosocial treatments (Sonuga-Barke et al., research suggesting the foundational value of posi-
2013). In addition to incorporation of common tive early dyadic social interaction for later child
blinded outcomes, a further key next step in social communication and adaptation (Landry,
methodological development will be widespread Smith, Swank, & Miller-Loncar, 2000; Page, Wil-
implementation of pre-recruitment trial registration helm, Gamble, & Card, 2010; Tomasello, 2008).
and pre-specification of analysis plans and primary Alongside this, autism developmental science has
outcomes, avoiding the post hoc selective reporting identified variations from such normative patterns in
© 2018 The Authors. Journal of Child Psychology and Psychiatry published by John Wiley & Sons Ltd on behalf of Association for
Child and Adolescent Mental Health.
426 Jonathan Green and Shruti Garg J Child Psychol Psychiatr 2018; 59(4): 424–43
young children at risk of or diagnosed with the (Mahoney, Powell, & Finger, 1986), Manchester
condition. Thus increased parent directiveness (but Assessment of Caregiver Infant interaction (MACI;
not reduced responsiveness) is found in the latter (Wan, Brooks, Green, Abel, & Elmadih, 2017), Emo-
part of the first year in infant siblings at risk of tional Availability Scale (EAS; Biringen, Derscheid,
autism (Harker, Ibanez, Nguyen, Messinger, & Vliegen, Closson, & Easterbrooks, 2014), but also
Stone, 2016; Wan et al., 2013) along with reduced with event-sampling (Kasari, Siller, et al., 2014).
child attentiveness to parent, affective signalling and Two recent studies in the French and Kennedy
coordinated dyadic communication (Parlade & Iver- (2018) review found treatment effects on parent
son, 2015; Wan et al., 2013). Similar patterns are ‘responsiveness’ in autism. A 12 week ‘Focused Play-
seen in established autism as well as in other time Intervention’ (FPI) that combined live modelling
developmental disabilities (Blacher, Baker, & Kalad- from therapist to parent with video-feedback and
jian, 2013; Doussard-Roosevelt, Joe, Bazhenova, & psycho-education was tested in a pre-emptive study of
Porges, 2003). We cannot of course assume that 66 toddlers at high autism-risk mean age 22 months
such patterns of PCI are responsible for autism (Kasari, Siller et al., 2014). It found a positive end-
outcomes; it could be in theory that these different point effect on level of parental responsiveness, but
PCI trajectories are incidental or even adaptive the effect was not sustained at 3 months follow up nor
(Mottron, 2017). Nevertheless, a transactional associated with effect on child variables (language
account (Sameroff, 2009) of the early reciprocal scores, joint attention or child communication, diag-
interplay between intrinsic developmental atypicali- nostic outcome). A 10-week Joint attention, Symbolic
ties in autism and the early social environment Play, Engagement and Regulation (JASPER) interven-
would suggest that evoked perturbations in interaction also produced proximal effect on responsiveness
tion can indeed serve to amplify social impairments in an RCT of 86 toddlers with diagnosed autism
and symptom impact over time; on this basis, spectrum disorders (ASD) (22–36 months) against
reversing these deficits and optimising psycho-education, here associated with improved
(‘normalising’) social interaction could in theory child-initiated joint engagement with parent (Kasari,
improve social functioning or reduce symptoms in Gulsrud, Paparella, Hellemann, & Berry, 2015; Shire
autistic children. The PCI target has the added et al., 2017). In contrast, a parent-mediated version of
theoretical advantage of being ‘naturalistic’ (i.e. the behaviour-learning focused Early Start Denver
working with the everyday developmental/family Model (ESDM) using therapist direct coaching and
system of the child) and potentially more easily modelling for parents in 98 children in a 12 week wait-
generalisable than clinic-based interventions list controlled trial, found no effect on responsiveness
directly with the child. It is however also based on or other parent interaction or child outcome (Rogers
the assumption that intervening in the developmen- et al., 2012). Initial report of the subsequent 3 site
tal system in this way will be ‘powerful’ enough trial of ESDM on this cohort also found no overall
significantly to impact autism development, tradi- group effect (Rogers et al., 2014), so prior parent
tionally seen to be intrinsic and relatively environ- coaching did not appear to enhance subsequent
mentally stable. Whether this assumption holds is therapist-mediated treatment.
an empirical question to which we will return. Studies of interventions using other methods have
Interventions designed to impact PCI have often found effects on non-directiveness but not respon-
used demonstration and modelling by the therapist siveness. A 3 month six session video-feedback
working directly with the child in the parent’s intervention (VIPP-AUTI) tested against home-based
presence or mixed with didactic education. More nursing care in 78 children with ASD aged 16–
recently, there has been a rise in more exclusive 61 months (Poslawsky et al., 2015) found a treat-
work with the parent using direct coaching in real- ment effect on reduced parental ‘intrusiveness’ in
time during interaction or use of video-feedback dyadic interaction at endpoint but no effect on
techniques. The precise ‘proximal’ intervention tar- parent sensitivity or structuring. Child effects on
gets within PCI (that is, the immediate goal for initiation of joint attention (but not other aspects of
change with intervention) have varied from parental behaviour) were found at 3-month follow up,
‘responsiveness’ and ‘non-directiveness’, parental although not associated with the prior parent
‘communicative synchrony’, to mutual ‘joint atten- change. A separate 12-session ‘iBASIS-VIPP’ inter-
tion and engagement’ and ‘affect sharing’ (see Fig- vention was tested in a pre-emptive RCT for infants
ure 1). Each of these targets implies a slightly at familial autism-risk against no intervention in 54
different theoretical orientation and defined goals. dyads, and found to have good effect at 15-month
endpoint in reducing parent non-directiveness, but
Parental responsiveness and non-directive- no effect on responsiveness; there was associated
ness. The concept of parental responsiveness is improvement in child attentiveness to parent (Green
often associated with early relational research, but et al., 2015). Two-year follow up of this trial (Green
terms are used subtly differently between groups. et al., 2017) showed sustained improvement in par-
These aspects of PCI are typically coded with global ent non-directiveness 1 year after treatment but
measures such as Maternal Behavior Rating Scale evidence of fall-off thereafter, although improvement
Figure 1 Schematic figure of proximal and more distal intervention and mechanism targets in autism psychosocial intervention (see text)
[Colour figure can be viewed at wileyonlinelibrary.com]
in child communication initiation with parent and

Joint engagement and Joint attention. An alter-
prodromal symptoms was sustained (see below).
native target for treatment is the quality of shared
interaction rather than the separate behaviour of
Parent communicative synchrony. The concept of
each partner. Typically, this will be measured as
synchrony is allied to ‘responsiveness’ but derived
‘duration’ counts of the phenomenon of interest from
more from the communication development litera-
video-tape and like all these methods importantly
ture, predicting communication and language out-
can be coded blind to treatment allocation. ‘Joint
comes in autism (Siller & Sigman, 2002) and usually
engagement’ as an outcome is a particular feature of
measured using event-sampling rather than global
the JASPER model and a number of trials have
rating. Subtle differences in the way this behaviour
shown reliable increases in joint engagement or joint
is conceptualised and measured again hamper com-
attention following relatively brief (usually 12 week)
parison across treatments at a detailed mechanism
treatments. Typically, the caregivers are coached in
level. Thus, one definition is of ‘parental verbal
the treatment model, encouraged to use strategies
behaviour directed to the child’s focus of attention
for setting up the learning environment, modelling
and actions’ (i.e. child-focused behaviours) (Siller,
and prompting for joint attention, expanding play,
Hutman, & Sigman, 2013), and another is of ‘parent
and using developmentally appropriate language.
communication responses that acknowledge, con-
For instance, Kasari, Lawton et al. (2014); ran-
firm or reinforce the child’s focus, play, actions,
domised 147 ‘low-resourced’ families of a child
thoughts or intentions; in tune with what the child is
diagnosed with ASD, aged 24–60 months, to either
thinking, saying or doing rather than redirecting the
a caregiver-mediated intervention (CMM) following
child away from his play, thoughts or communica-
the 12-week JASPER treatment and involving the
tion’ (Aldred, Green, Emsley, & McConachie, 2012).
child, or a group caregiver-only education pro-
Parental synchrony has been targeted particularly
gramme. Significant interaction between treatment
by the video-feedback Preschool Autism Communi-
group and time during treatment was identified on
cation (PACT) treatment, which targets social inter-
the primary outcome of Joint Engagement, with the
action and communication impairments, first by
caregiver-mediated group exhibiting a significantly
increasing parental sensitivity to child communica-
greater rate of improvement. Both groups also
tion and reducing mistimed responses using video-
demonstrated improvement in joint attention, but
feedback, and second by promoting a range of
not in functional play. Chang, Shire, Shih, Gelfand,
positive social communication strategies. PACT was
& Kasari (2016) randomly assigned 66 children with
tested in an RCT of 152 children with core autism
ASD aged 36–60 months to a teacher-delivered
(age 2–5 years), which found large endpoint effect in
JASPER intervention or wait-list control. Endpoint
optimising parental synchrony across three geo-
intervention effects were reported for a number of the
graphically varied contexts in the United Kingdom
multiple measured outcomes, including child-
(Green et al., 2010; Pickles et al., 2016). The same
initiated joint engagement and joint attention, recep-
effect was replicated in an RCT testing adaptation of
tive language, but not on others such as expressive
PACT to South Asia in 64 children aged 2–9 years
language or symbolic play. However, a qualification
(Rahman et al., 2016). The FPI intervention as
would have to be that this interpretation did not
above, also partially video-aided, found small gains
adjust for the multiple testing. A third study (Landa,
in parental synchrony confined to dyads with more
Holman, O’Neill, & Stuart, 2011) randomly assigned
language-delayed children (Siller et al., 2013).
50 children, aged 21–33 months with a diagnosis of prompts had no effect. The pattern of this result
ASD, to parent education and parent education could, however, be subject to ‘training to the test’;
supplemented with a group focusing on ‘Interper- parental contingent reinforcement of child utter-
sonal Synchrony’ within a supplementary social ances, trained in the therapy, produced local con-
curriculum. Intervention took place for two-and-a- text-specific increase in the target child behaviour.
half hours a day, 4 days per week, for 6 months. A Evidence for generalisation is modest; independent
significant treatment effect was found for socially global impression of social communication from
engaged imitation, with a large increase (17%–42%) same tapes alongside parent interview reported
in imitated acts paired with eye contact in the improvements as did parent-reported communica-
Interpersonal Synchrony group (Landa et al., tion on Vineland, but there was no effect in parent-
2011). A caveat is that these outcomes were collected reported language use, standard language measures
within the training environment and close to the or parent-rated autism behaviours. A subsequent
trained targets of intervention. No significant similar intervention model combining PRT and ABA
between-group differences were observed for initia- reported effects on dyadic joint engagement and
tion of joint attention and shared positive affect, shared enjoyment but also no generalised effect
although trends were seen. (Brian, Smith, Zwaigenbaum, & Bryson, 2017).
Behavioural training methods Summary - proximal dyadic targets

Behavioural learning methods use focused beha- The evidence outlined above shows that caregiver
viour modification techniques to influence behaviour dyadic behaviours with their child (Figure 1A) are
change. In traditional behavioural interventions for notably responsive to a number of different inter-
autism the targets are reduction in unwanted (often ventions (just as they can be in non-autism con-
repetitive or assumed non-socially functional) beha- texts). This applies across the spectrum of social
viours along with increase in identified social class, ethnicity and education. Replicated effects
behaviours. In more recent studies, the targets for from robust trials of various interventions in autism
behavioural change are more developmentally show moderate to large effect sizes, particularly to
informed; for instance, in pivotal response therapy improve parental responsiveness sensitivity and
(PRT) the targets are behaviours such as motivation synchrony, theoretically linked in development to
for engagement and interaction considered theoret- positive child social and communication outcomes.
ically ‘pivotal’ for generating wider generalised In relation to intervention techniques, video feed-
change. Broader ‘curriculum’ models such as Early back probably has the strongest evidence for pro-
Start Denver Model (ESDM) select a range of devel- ducing reliable parental change in this kind in both
opmentally relevant targets, from communication to autism (Aldred, Green, & Adams, 2004; Green et al.,
play to motor activity. What distinguishes these 2010; Kasari, Siller et al., 2014; Poslawsky et al.,
models however is that the methods used to achieve 2015; Rahman et al., 2016) and neurotypical groups
therapeutic change towards such targets are still (Juffer, Bakermans-Kranenburg, & van IJzendoorn,
taken from behavioural learning theory, largely 2008). Techniques primarily using group or individ-
using repetition and contingency reinforcement for ual coaching and modelling techniques with parents
desired behaviours (Koegel, Singh, Koegel, Hollings- do not show such reliable results (Carter et al.,
worth, & Bradshaw, 2014). While use of contingen- 2011; Rogers et al., 2012). Targets involving recip-
cies is part of naturalistic parenting, its use in these rocal dyadic interaction between parent and child,
models is much more exclusive and intensively such as joint engagement and joint attention (Fig-
focused. Interventions of this kind are typically ure 1B) have also proved treatable in replicated
intensive and therapist-child delivered over several studies. There is less consistent evidence that the
years. The only trial of this type of intervention child’s dyadic social interaction in turn can be
identified by French and Kennedy (2018) as having improved (Figure 1C). Many studies demonstrating
acceptable low risk of bias was, however, of a increased parental responsiveness have not found
relatively low intensity (one session/week over that this transmitted well into child effects; others
12 weeks) group PRT intervention, focused on have shown change in both parental responsiveness
improving functional communication (Hardan et al., and child-initiated joint engagement joint attention
2015). The primary outcome was the frequency of (Shire et al., 2017). However, focused video-feedback
functional child utterances in the dyad coded during for parents in the PACT and iBASIS trials and a mixed
structured lab interaction in which parents were coaching, modelling and video work in JASPER are
instructed to ‘try getting the child to communicate as all associated with reciprocal effects on spontaneous
much as possible’. Parents were rated as implement- child communication in the treatment dyad.
ing PRT methods successfully; total child utterances Convergent evidence thus demonstrates that
increased in the PRT group, with the effect driven by immediate PCI targets are amenable to intervention.
imitative and non-verbally prompted utterances However, such targets are highly proximal to the
rather than spontaneous utterances; and verbal intervention effort, and in many studies these
outcomes are measured within or in very similar Kuhfeld, Kasari, & McCracken, 2017). Intervention
contexts to the treatment setting, potentially trials highlighted by French and Kennedy (2018)
confounding independence and making possible a reporting concurrent autism symptom outcomes
simple ‘training to the test’ result. It is moreover a have had mixed results. Dawson et al., 2010;
feature of autism development that the generalisa- showed no effect on ADOS CSS scores after a 2 year
tion of acquired skills across contexts is particularly Early Start Denver Model (ESDM) intervention and
difficult: it cannot be assumed that short-term Fletcher-Watson (Fletcher-Watson et al., 2016)
changes in dyadic interactions of this kind will found no effect on BOSCC symptoms with a com-
translate well into more generalised social gains for puter-aided iPad intervention. No parent-rated SRS
the child. In this sense, the majority of autism effect was observed with PRT treatment (Hardan
intervention science to date (and the vast majority of et al., 2015). On the other hand, significant endpoint
published papers) has not really moved beyond what reduction in total symptom severity on the ADOS
are in effect the equivalent of pre-clinical or early CSS was found after the 1 year preschool PACT
phase studies looking at potential developmental intervention (Pickles et al., 2016) where change in
mechanisms, on the (implicit or explicit) assumption Social Communication symptoms alone had failed to
that these will go on to effect more generalised autism reach significance (Green et al., 2010). It has been
development and functioning. Even if well done, we harder to show effects on other concurrent outcomes
can hardly infer from such evidence that a treatment using objective measures. For instance, 2-year
is ‘effective for autism’: and the (frequent) claims to intensive ESDM training tested against usual care
this effect are inflated and unjustified. in 48 children (Dawson et al., 2010) reported signif-
icant improvements in IQ (this was carried by
improvements in IQ components of receptive and
Concurrent child outcomes beyond the dyad
expressive language), but not other child measures.
By contrast, measuring child concurrent outcomes The IQ effects were not sustained at 2-year follow-up
beyond the immediate dyadic context (Figure 1D) (Estes et al., 2015).
represents a first stage of true generalisation for
interventions whose proximal focus is the dyad. It
Downstream developmental targets – symptoms
also represents the immediate outcome for therapist-
and adaptation
child interventions without parental involvement. An
example of such concurrent child social outcome is There has been a relative lack of developmental
the behavioural symptoms that define autism disor- follow-on studies after early intervention (Figure 1E),
der itself. although the situation is now beginning to change.
The results are somewhat encouraging but have
Autism symptoms. The symptoms of autism are been mixed. On relatively short-term follow-up,
most rigorously measured in standardised Poslawsky et al. (2015) found sustained treatment
researcher administered tests of social interaction, effect on child initiation of joint attention at
which are reliable and can be blind-coded. Because 3 months after the video feedback VIPP-AUTI inter-
these symptom measures capture the child’s social vention, but not on other measures. Kasari, Papar-
communication with an unfamiliar adult in a context ella, Freeman, & Jahromi, 2008 showed effect on
different to the parent-child dyad or treatment set- objective language outcomes from the JASPER inter-
ting, they do represent the generalisation of any vention using the Reynell Developmental Language
treatment gains across both context and interaction Scales (RDLS) at 12-month follow-up but this was
partner. The most validated of these symptom mea- not replicated in a later study (Kaale, Fagerland,
sures is the Autism Diagnostic Observation Sched- Martinsen, & Smith, 2014). Chang et al., 2016
ule (ADOS-2 (Lord et al., 2000), which can rate social measured uncontrolled follow-up of the intervention
communication (SA) and repetitive and restrictive group only and reported mixed findings. Kasari,
behaviour (RRB) dimensions separately or together Lawton et al. showed maintenance of improvement
in a ‘Combined Severity Scale’ (ADOS CSS). A new in their CMM group at 3-month follow-up, but no
related measure from the ADOS originators is the between-group effect. Other controlled studies that
Brief Observation of Social Communication Change showed between-group endpoint effects on either
(Grzadzinski et al., 2016) designed to be more sen- parent or child dyadic social behaviours do not see
sitive to total symptom change in treatment trials. them at short or medium-term follow-up (Baranek
Parent-rated symptom measures such as the Social et al., 2015; Kasari, Siller et al., 2014).
Communication Questionnaire (Rutter, Bailey, & Four RCTs have however now reported longer-term
Lord, 2003) and the Social Responsiveness Scale (>1 year) autism symptom outcomes after preschool
(Constantino, 2002) are commonly used but are less intervention. Estes et al. (2015) measured ADOS
rigorous because they are unblinded. The specificity total score at a mean age of 6 years, 2 years after the
of the SRS has also been significantly challenged end of a 2-year intensive developmental behavioural
since it measures such a broad range of behaviours intervention (ESDM). While there had been no effect
beyond core autism (Kaat & Farmer, 2017; Sturm, on symptoms at trial endpoint, the 21 of 24 children
in the intervention group followed up showed greater the measurement error intrinsic to observational
total and RRB symptom reduction over the follow-up video coding. Treatment effect on parent synchrony
period compared to 18 of 24 followed from the strongly mediated (90%) the change found in child
control group. Causal treatment effect inference communication initiations with parent, supporting
from these findings is limited however by the lack the treatment theory. Both these mediation analyses
of intention to treat (ITT) analysis or adjustment for are thus convergent in identifying similar specific
baseline pretreatment variables. Kaale et al. (2014) parental interaction behaviours (‘mirrored pacing’
reported no effect on non-blinded parent- or teacher- and ‘communicative synchrony’) that improve child
rated autism symptoms (SCQ) 12 months after a 12- engagement and communication in the dyadic con-
week teacher-mediated JASPER intervention text. They stand as the clearest mechanism findings
(n = 33) compared to controls (n = 27). In contrast, to date in autism intervention science and provide
follow-up of the original PACT trial, 6 years after important and generalisable information about what
treatment endpoint, assessed 121 of the original 152 are the active components of the therapies in facil-
trial participants with core autism at mean age itating interaction and child outcomes. Furthermore,
10.5 years. ITT analysis showed reduction in autism the Pickles et al. analysis also showed that it was the
symptom severity (ADOS CSS) at both treatment change in child dyadic communication initiations
endpoint (ES 0.64, 95% CI 0.07, 1.20) and follow-up with parent (rather than the parental synchrony)
(ES 0.70, 95% CI 0.05, 1.47), resulting in a that mediated the subsequent change in child
moderate averaged treatment effect on symptoms autism symptoms with a researcher – thereby
over the total period (ES 0.55, 95% CI 0.14, 0.91). demonstrating a causal chain of generalisation of
Non-blind parent-rated autism symptoms on SCQ child-acquired skill from the PCI context into a
(ES 0.40, 95% CI 0.05, 0.77) and repetitive beha- researcher-child interaction. More mechanism tests
viours on RBQ (ES 0.87, 95% CI 0.47, 1.35) also of this kind will be important going forward.
showed comparable improvement at follow up (Pick-
les et al., 2016). A similar video-feedback social
Summary of psychosocial mechanism targets
communication intervention for infants at-risk of
autism, implemented between 9 and 14 months of Autism particularly impacts social functioning in
age and followed at one and then 2 years from end of development. We have seen above that a transac-
treatment (Green et al., 2017), showed interestingly tional account of social development and reciprocity
parallel ITT results, with a significant effect of in autism, recruiting ideas from social-developmen-
intervention over the treatment and follow-up period tal research in neurotypical populations, has led to a
for autism prodromal symptoms (ES = 0.32; 95% CI rationale for interventions targeting key aspects of
0.04, 0.60; p = .026), parental dyadic non-directive- social development in autism. The evidence to date
ness/synchrony (ES = 0.33; CI 0.04, 0.63; p = .013), shows that a number of such theoretically based and
and child attentiveness/communication initiation targeted interventions can indeed reliably improve
(ES = 0.36; 95% CI 0.04, 0.68; p = .015). various aspects of the local dyadic social communi-
cation of children with autism, particularly with their
caregivers (Figure 1A–C, Table 1). There is less work
Mediation testing in psychosocial intervention
on whether such dyadic change can produce more
Only two studies in the extant autism literature have generalised impact on concurrent within-child out-
mounted a substantive test of mechanism through a comes (Figure 1D): there have been positive findings,
mediation test within an RCT. Gulsrud, Hellemann, but the overall evidence is mixed. There is even less
Shire, & Kasari (2016) used videotape analysis of the work on the crucial issue of long-term developmental
content of therapeutic sessions in the JASPER outcomes (Figure 1E), but some emerging evidence,
intervention on key putative active ingredients. especially from the PACT trial, that this is possible
There was a possible methods confound in that over the medium to long term. With a developmental
mediator and primary outcome (joint engagement) disorder like autism, interventions must surely
were both coded (albeit independently) at endpoint aspire to such long-term effects in order to be
from the same video-material with no midpoint convincing as disorder-specific treatments.
assessment, allowing the possibility of ‘reverse cau- Given the convergent evidence that short-term
sation’ in the inference of effect. Nevertheless, the context-specific proximal change is possible using
analysis did identify a particular aspect of the a variety of intervention methods, further replication
therapy (parental ‘mirrored pacing’) as mediating of this established result on its own is redundant
the joint engagement outcome; which is an impor- moving forward (although they are the easiest and
tant result with generalisable value and points cheapest designs to carry out!). It is true that
forward to potential further refinements of the ther- mechanism findings from one trial (Pickles et al.,
apy procedure. Within the PACT trial, Pickles et al. 2015) show a causal link between treatment change
(2015), identified a latent midpoint mediator variable on child dyadic interaction with caregiver and gen-
using baseline, midpoint and endpoint repeated eralised symptom change across context, and this is
measures, a method of analysis that also reduces encouraging that these proximal changes may
Table 1 Summary of psychosocial mechanism targets and intervention outcomes (see text)
Target Definition and measurement Intervention effects
Parent dyadic behaviours

Parent responsiveness/ Global rating of sensitive No change with video-feedback (Green, 2015;
sensitivity responding (EAS, MACI, MBRS). Poslawsky et al., 2015) or behavioural modelling
Event count (Rogers et al., 2012).
(Kasari, Siller 2014) Improved on focused playtime intervention (parent
coaching and video-modelling) not sustained
on FU (Kasari, Siller et al., 2014; Siller et al., 2013)
Parent intrusiveness/ Reduced intrusiveness (Green et al., 2015;
directiveness Poslawsky et al., 2015)
Parent communicative ‘Child focused dyadic behaviours’ Improved on video-aided intervention (Green et al., 2015;
synchrony (Siller et al., 2013). DCMA Rahman et al., 2016; Siller et al., 2013)
Improved and mediates child initiation change on
video-aided intervention (Green et al., 2010;
Pickles et al., 2015)
Child dyadic behaviours
Child dyadic MACI/DCMA Improved in PACT, sustained 6 years after treatment
communication/ end (Pickles et al., 2016).
attentiveness with Improved in iBASIS, sustained 2 years after treatment
parent end (Green et al., 2017).
Improved initiation of JE language and requests in
JASPER (Chang et al., 2016)
Child initiation of joint Video coding Effect on child imitation but not on IJA
attention (IJA)/socially (Landa et al., 2011).
engaged imitation Improved prompted ‘child utterances’ with parent
(Hardan et al., 2015)
Reciprocal dyadic behaviours
Dyadic joint engagement, Joint engagement (JE) measure JE improved with JASPER (Kasari 2008, Kaale et al., 2014)
joint attention, joint play Increased joint engagement and play (Shire et al., 2017)
No gains in shared affect (Landa et al., 2011).
Short term improved emotional engagement but no
follow up effect (Brian et al., 2017).
No effect on play with parent (Poslawsky et al., 2015)
Improved Clinician Global Impression (Hardan et al., 2015)
Child outcomes beyond the dyad
Adaptive function Parent-rated Effect in Harden et al., 2015
Adaptive outcome VABS
Social and Research rated ESCS Improved initiation of joint attention at 3 month FU
Communicative Parent rated Social and (Poslawsky et al., 2015)
skills communicative skills CSBS Improved parent CSBS (Green et al., 2010)
Structural language Research rated (PLS, Mullen) No effect on objective or parent rating (Hardan et al., 2015).
Parent-rated MCDI No effect on objective rating but effect on receptive
language on parent rating (Green et al., 2010).
Verbal IQ effects from ESDM Dawson et al., 2010 but not on
FU (Estes et al., 2015)
Autism Symptoms Research rated (AOSI, No ADOS effect from ESDM Dawson et al., 2010 or SA
ADOS, BOSCC) symptoms from PACT (Green et al., 2010). No BOSCC
Parent-rated (SRS) effect from iPAD intervention (Fletcher-Watson et al.,
2016) or PRT (SRS) (Hardan et al., 2015).
Effect on ADOS CSS (Pickles et al., 2016)
Child outcomes generalised over time
Structural language Research rated (PLS, Mullen, RDLS) Effect 12 months post treatment (Kasari et al., 2008)
Parent-rated MCDI not replicated (Kaale et al., 2014)
Language composite Language composite no effect (Pickles et al., 2016)
Autism Symptoms Research rated (AOSI, Effect at 2 year FU but not ITT (Estes et al., 2015)
ADOS, BOSCC) Symptom effect (AOSI/ADOS) over 2 year post treatment
Parent-rated (SCQ, RBQ) (Green et al., 2017)
Symptom effect (ADOS CSS, SCQ, RBQ) over 6 yrs
post treatment
(Pickles et al., 2016)
No effect from JASPER on SRS (Kaale et al., 2014)
SRS, Social Responsiveness Scale; SCQ, Social Communication Questionnaire; VABS, Vineland Adaptive Behaviour Scales; MCDI,
Macarthur-Bates Communicative Development Inventory; PLS, Preschool Language Scales; CSBS, Communication and Symbolic
Behaviour Scales; AOSI, Autism Observation Scale for Infants; ADOS CSS, Autism Diagnostic Observation Schedule Comparative
Severity Score; BOSCC, Brief Observation of Social Behaviour Change; RDLS, Reynell Developmental Language Scale; RBQ, Repetitive
Behaviour Questionnaire; ESCS, Early Social Communication Scales; DCMA, Dyadic Communication Measure for Autism; EAS,
Emotional Availability Scales; MACI, Manchester Assessment of Caregiver-Infant Interaction; MBRS, Maternal Behaviour Rating Scale.
indeed be meaningful in relation to symptoms and demonstrated the searched-for causal chain between
generalised adaptation. But such a finding needs genetic variation, cell biology, neural system atypi-
replication, and the field now needs larger and more cality and phenotypic behaviour. Thus for instance
ambitious psychosocial trials that address in animal knock-out models of the monogenic disor-
mechanism, generalisation and the possibilities of der neurofibromatosis 1 (NF1), pharmacological
long-term developmental and adaptive change, as intervention or genetic manipulation can downregu-
well as adaptive interventions tailored to patient late a cell signalling abnormality directly related to
characteristics and individual response (for instance the genetic variation and show consequent effects on
using sequential multiple assignment randomised improved synaptic function, synaptic protein expres-
‘SMART’ trials). Accumulation of such trial evidence sion, long-term potentiation and finally rescue of the
will refine approaches and further answer an endur- cognitive and behavioural phenotype (Li et al., 2005;
ing question; what are the limits to clinically relevant Molosh et al., 2014). Analogous results have been
effectiveness of ‘environmental’ interventions in a shown in other syndomic autism conditions such as
condition that is highly heritable and developmen- fragile X (FXS) and tuberous sclerosis complex (TSC).
tally persistent? Such animal findings will not necessarily translate
into human study (see below) but they provide a
proof of principle that intervention in theoretically
Moving to the level of neural mechanism relevant neural system targets can produce pre-
The neurodevelopmental context of autism inevitably dictable and relevant improvements in relevant
leads to a consideration of expanded targets for behaviour outcomes. This has led to an accelerating
treatment beyond the purely psychosocial into the effort to study the effect of intervention in humans on
assumed neural system pathogenesis. No straight- a variety of these neural system targets relevant to
forward targets present themselves. The known autism, many but not all of which involve work with
genetic background of autism shows great hetero- monogenic syndromic autism. We thus now review
geneity and no common final pathways of effect the various neural system mechanisms targeted in
within the neural system linked to the behaviour these studies, moving in turn from the most ‘prox-
phenotype have yet been established. At a descrip- imal’ in terms of synaptic function; to post-synaptic
tive level, the large number of genetic variants molecular pathways; to more general brain-growth
associated with autism outcomes show convergence and neurotropic factors; and finally, to more gener-
on functional networks related to intracellular sig- alised neural system function reflected in EEG and
nalling (see below), neuronal development and axon connectivity measures (see Figure 2).
guidance, and chromatin modification and tran-
scription regulation (Pinto et al., 2014), but brain
Neurotransmitters/neuromodulatory system targets
localisation of such effects is less established and
other findings point to generalised functional Glutamate. Glutamate is the main excitatory neu-
aspects such as neural system connectivity rotransmitter in the brain, exerting its effect through
(O’Reilly, Lewis, & Elsabbagh, 2017). Further chal- the activation of several receptor subtypes. Fast
lenges arise in linking such neuroscience findings to excitatory synaptic transmission is mediated via
the autism cognitive and behavioural phenotype. ionotropic (AMPA, NMDA and kainite) receptors,
Prospective neuroscientific studies of the emergence whilst modulatory action is exerted through metabo-
of autism in the prodrome have suggested some tropic G-protein coupled (mGluRs) receptors. The
specific risk markers, but as the work has gone on mGluRs play an important role in synaptic plasticity,
more and more aspects of early neurodevelopment learning and memory and glutamate signalling is
are found to be atypical in autism emergence (John- hypothesised to be relevant to aetiology of chronic
son, Gliga, Jones, & Charman, 2015; Szatmari et al., brain disorders such as schizophrenia and Alzhei-
2016), replicating perhaps a sense of the hetero- mers dementia as well as ASD (Uzunova, Hollander,
geneity in other aspects of autism science. & Shepherd, 2014). FXS animal model work sug-
A promising strategy for limiting this heterogeneity gests that exaggerated mGluR signalling induces
has been to constrain the primary genetic variance enhanced hippocampal long-term depression related
by studying monogenic disorders with a high autism to the behavioural phenotype (Bear, Huber, & War-
expression. As well as simplifying the biological ren, 2004). Fenobam, the first mGluR antagonist
system for study, this strategy has the added used in human clinical trials, showed reduced
advantage that knockout animal models can be bred anxiety, hyperarousal and attention in FXS patients
for detailed biological investigation. There is an in a small (n = 12) single-dose, single-arm open label
inevitable question of whether these ‘syndromic study (Berry-Kravis et al., 2009), but no Phase II
models’ of ASD in humans actually represent the studies have been reported, probably due to CNS
same phenotype as the idiopathic condition; detailed adverse effects with Fenobam use (Freidmann,
phenotypic studies suggest they can do (Garg et al., Davis, Ciccone, & Rubin, 1980). Mavoglurant
2015), but this remains an important question. (AFQ056), a structurally novel noncompetitive
Some of these syndromic models have indeed mGluR5 inhibitor, showed promise in FXS mouse
Neurotransmitters
Post synaptic receptors
Intracellular signalling
Pathways
Figure 2 Simplified representation of key biological pathways and therapeutic targets commonly implicated in syndromic autism (see
text). Abbreviations: mGluR, metabotropic glutamate receptors; NMDAR, NMDA receptor; AMPAR, AMPA receptor; NF1, Neurofibro-
matosis type 1; TSC, Tuberous Sclerosis Complex; PTEN, Phosphate & tensin homologue; FMRP, Ras; ERK, extracellular signal regulated
kinase; MEK, mitogenactivated protein kinase, FMRP, fragile X mental retardation protein; mTOR, mammalian Target of Rapamyci; PI3,
phosphatidylinositol 3-kinase; IGF, insulin like growth factor; BDNF, brain derived neurotrophic factor
models in rescuing the aberrant dendritic spine SRS and ADHD rating scales (Erickson, Mullett, &
morphology and concurrent improvements in the McDougle, 2009).
molecular and social behavioural phenotype; how-
ever, results from human clinical trials have been GABA. GABA is the main inhibitory neurotrans-
less encouraging. Two 12-week, multi-centre phase mitter in the brain, exerting its effect through the
2b RCTs of Mavoglurant in 175 adults and 139 GABAA and GABAB postsynaptic receptors. Abnor-
adolescents with FXS, showed no significant treat- mal GABAergic signalling has been demonstrated in
ment effects between mavoglurant and placebo on a number of syndromic autism models such as FXS
primary endpoints of FXS specific Aberrant Beha- and NF1. The leading hypothesis here is that abnor-
viour Checklist (ABC) (Berry-Kravis et al., 2016). mal GABA signalling leads to an inhibition/excita-
These results suggest that manipulation of the tion imbalance, including dysregulation in protein
mGluR signalling alone may be insufficient for synthesis. GABAergic signalling is now consequently
attenuating the FXS autism phenotype in humans. an important therapeutic target in ASD. In FXS,
Defects in NMDA receptors have been reported in along with increase in glutaminergic signalling, there
FXS and Rett syndromes although the mechanism is is deficiency in GABA receptor expression and GABA
not clearly understood. Memantine, an NMDA recep- mediated inhibition, which has led to the develop-
tor antagonist approved for use in Alzheimers dis- ment of therapies aimed to increase GABA mediated
ease, improved attentional processes and working inhibition. A double-blind RCT of GABAB agonist
memory in patients with Fragile X-associated tre- arbaclofen in 63 FXS patients aged 6–40 years found
mor/ataxia syndrome (Yang et al., 2016). In FXS, a significant treatment effects in social behaviours and
small open label trial of memantine in six subjects, function (Berry-Kravis et al., 2012) but not on the
mean age 18.3 years showed symptom improvement primary endpoint measure of the Aberrant Beha-
on Clinical Global Impressions (CGI) but no effect on viour Checklist-irritability subscale. A recent dou-
symptom specific rating scales, which included ABC, ble-blind placebo controlled RCT of Arbaclofen of
125 adolescents/adults (age 12–50 years) and 172 subjects (Veenstra-Vanderweele et al., 2009); and
children (age 5–11 years) found no significant treat- mutations in serotonin transporter genes such as
ment effect in either group on the primary outcome SLC29A4 have been linked to ASD symptoms
measure of FXS specific ABC social avoidance sub- (Adamsen et al., 2014). Neuroimaging studies using
scale. However, in the child study, significant effects PET have shown significantly lower levels of sero-
were noted on secondary outcome measures of ABC- tonin synthesis in ASD children compared to con-
irritability subscale and parenting stress index, trols (Chandana et al., 2005), particularly in medial
suggesting that younger patients may derive some frontal cortex, midbrain and temporal lobe areas
benefit (Berry-Kravis et al., 2017). Acamprosate is a (Makkonen, Riikonen, Kokki, Airaksinen, & Kuikka,
mGluR5 receptor antagonist, a GABAA receptor 2008). Treatment with selective serotonin uptake
agonist, and has complex actions on NMDA recep- inhibitors has been shown to have some effect in
tors depending on glutamate concentration (Erick- ameliorating the repetitive/obsessive behaviours in
son, Early, et al., 2011). A small open label trial in 12 ASD; however, a Cochrane review of nine RCTs
FXS children aged 6–17 years suggested some found no overall evidence of positive effect of SSRIs
improvements in social behaviours, inattention and on core features (Williams, Brignell, Randall, Silove,
hyperactivity (Erickson, Mullett, & McDougle, 2010); & Hazell, 2013).
larger studies using Acamprosate in FXS are cur-
rently underway and results from trials of other Oxytocin. Oxytocin is widely distributed in the CNS
GABA agonists such as ganaxolone are awaited. and plays an important role in social recognition,
More recently, there has been interest in the diuretic memory, attachment, as well as stereotyped beha-
Bumetanide, which enhances GABAergic inhibition. viours. It mediates the perinatal excitatory to inhibi-
In a multi-centre phase 2 RCT (n- = 88, aged 2– tory shift of GABA. A number of human studies have
18 years), Bumetanide use was associated with examined the effect of oxytocin on social cognition,
significant improvement on parent and clinician- with mixed results. In a double-blind placebo-
rated autism measures (Lemonnier et al., 2017); control within-subject design, application of intra-
replication is awaited. nasal Oxytocin was associated with attenuated
amygdala response to faces regardless of valence
Dopamine. Dopamine plays a crucial role in synap- (Domes et al., 2007). In a RCT of 15 adults with ASD,
tic plasticity, cognitive functioning and neuropsy- oxytocin was associated with significant reduction in
chiatric pathologies. Interest in the role of Dopamine repetitive behaviours in comparison to placebo (Hol-
was stimulated by the observation that Dopamine lander et al., 2003). However, a number of studies
blockers (anti-psychotics) were effective in treating have reported no significant improvement of oxytocin
comorbidities associated with autism such as treatment. Dadds et al. (2014) evaluated oxytocin
aggression, hyperactivity and self-injury (Nakamura intervention in a double blind RCT in 38 children
et al., 2010). Deficits in dopaminergic functioning (age 7–16 years) with ASD administered during
have been demonstrated in syndromic models such parent-child interaction training sessions over four
as FXS, NF1 and TSC. A double-blind cross-over trial consecutive days and found no treatment effect on
of methylphenidate in 15 children with FXS demon- social interactional skills or emotional recognition.
strated that compared to placebo, two-thirds of Similarly, a double blind RCT of Oxytocin nasal
patients responded well with improvements in social spray twice daily in 50 males with ASD (12–18 years)
skills, hyperactivity and attention (Hagerman, Mur- showed no effects on caregiver rated SRS and CGI
phy, & Wittenberger, 1988). An open-label trial of (Guastella et al., 2015). One hypothesis for the
12 weeks of the partial dopamine D2 agonist Arip- differences between studies is that individual differ-
iprazole in 12 subjects with FXS showed good ences in endogenous oxytocin levels may moderate
tolerance and significant improvement in irritability treatment response. Addressing this issue, a recent
(Erickson, Stigler, et al., 2011), though this is not a RCT of 32 children with ASD investigated the efficacy
core autism symptom. of 4-week intranasal oxytocin treatment measuring
pre- and post-treatment blood oxytocin levels. The
Serotonin. Serotonin (5-hydroxytrytamine, 5-HT) study found treatment effects on SRS social abilities
is a critical modulator of neuronal interaction that confined to those with low baseline oxytocin levels
supports diverse behaviours and physiological pro- (Parker et al., 2017).
cesses including social behaviour, sleep, affective
regulation, learning, memory and synaptic plastic-
Signalling pathways targets downstream of
ity. The serotonergic system was one of the first
neurotransmitter receptors
neurotransmitter systems to be investigated in the
pathophysiology of ASD. Several studies have The MAPK/ERK pathway. Dysregulation of the
reported elevated levels of platelet serotonin in a intracellular Ras/MAPK (mitogen activated protein
third of ASD samples (Cross et al., 2008); short- kinases) signalling pathway is considered a major
term dietary depletion of serotonin is linked to risk factor for ASD and prominent in functional
increase repetitive behaviours and anxiety in ASD networks of autism-related genes (Pinto et al., 2014).
The RAS/MAPK pathway is embedded in a network Hyperactivity of the mTOR pathway is associated
of other pathways including PI3K and mTOR (see with several syndromic autism models including
Figure 2). Mutations in a number of genes on the Ras TSC, NF1, PTEN, and FXS. Dysregulated mTOR
pathways have been shown to be associated with signalling leads to abnormal development of funda-
ASD (Garg et al., 2013, 2017). NF1 is directly asso- mental processes such as axonal and dendritic
ciated with Ras/MAPK pathway overactivity, causing morphology, synapse formation and cell growth,
downstream increase in GABA and impairments in which have been associated with an ASD phenotype
synaptic plasticity, making this pathway a rational in animal models (Sato, 2016). Rapamycin is a
target for intervention. Statins downregulate the Ras specific inhibitor of mTOR and in pre-clinical study
pathway through inhibiting farnesylisation and this has improved social behaviour in TSC (Sato et al.,
rescues the social and behavioural phenotype in NF1 2012), PTEN-related disorders (Zhou et al., 2009)
knock-out animal models (Li et al., 2005). Subse- and 15q11–13 duplication (Oguro-Ando et al.,
quent translational trials in human NF1 have shown 2015). In TSC, preliminary trials of mTOR inhibitors
mixed results. Improvements in verbal and non- Sirolimus improved executive function (Davies et al.,
verbal memory were reported within a 12-week 2011) and Everolimus reduced seizure frequency
phase 1 observational study of lovastatin in 23 and ASD related symptoms on quality of life mea-
children aged 10–17 years (Acosta et al., 2011) and sures (Krueger et al., 2013). Based on these initial
in a 14-week RCT of lovastatin in 44 10–50 years results, several more TSC RCTs are currently under-
olds (Bearden et al., 2016); an n = 7 case series from way testing mTOR pathway inhibitors for neurocog-
the former cohort showed evidence of treatment nitive and ASD related deficits.
normalising pseudoresting state functional connec-
tivity in areas of the default mode network (Chaber-
Growth & Neurotrophic factor targets
naud et al., 2012). Four days of high-dose (200 mg)
lovastatin improved synaptic plasticity and phasic Deletion/mutation of the SHANK3 gene in Phelan-
alertness, as measured with trans-cranial magnetic McDermid syndrome (PMS; also known as 22q13
stimulation, in a case-controlled study of 11 adults deletion syndrome) results in reduced expression of
with NF1 (Mainberger et al., 2013). Larger statin scaffolding proteins in the postsynaptic density of
trials, however, have found little effect on cognitive/ excitatory synapses, impairing glutaminergic trans-
behavioural outcomes. A 12-week double-blind, mission and synaptic plasticity (Moessner et al.,
placebo controlled RCT of simvastatin in 62 children 2007). PMS is associated with 0.5%–2.0% of all
with NF1 aged 8–16 found no group differences on ASD cases. Similarly, loss of function mutations of
the Rey complex figure test (Krab et al., 2008); the X-linked gene MECP2 in Rett syndrome affects
another RCT of simvastatin (84 children aged 8– the structure and function of synapses at the
16 years) found no improvements in cognitive defi- microcircuit level critical for synaptic transmission
cits or parent-reported behavioural problems (van and plasticity (Guy, Hendrich, Holmes, Martin, &
der Vaart et al., 2013); and a 16-week RCT of Bird, 2001). Recent studies suggest that Insulin-like
lovastatin in 146 8–15 year olds with NF1 found no growth factor-1 (IGF-1) can have a beneficial effect
improvements in paired associate learning (Payne on synaptic development by promoting neuronal cell
et al., 2016). In contrast, a recent pilot data-rich survival, synaptic maturation and synaptic plasticity
RCT (n = 30) of simvastatin in young children with (Bozdagi, Tavassoli, & Buxbaum, 2013). A pilot
NF1–autism (4.5–10.5 years) studied for the first placebo-controlled double blind RCT using Insulin-
time the full hypothesised pathway from peripheral like growth factor in nine children with PMS was
MAPK activity, through multiparametric imaging of associated with significant improvements in social
neural system function to autism-related behaviours impairments and repetitive behaviours (Kolevzon
(Stivaros et al., 2018). Indications of possible statin et al., 2014). In Rett syndrome, an open label phase
effects towards normalising function were found at 1 study of IGF-1 in n = 12 girls with MECP2 showed
all these levels, encouraging the view that detailed good safety, tolerability and improvements in beha-
mechanism studies of this kind may have potential vioural abnormalities (Khwaja et al., 2014). Studies
to be illuminating, but the study needs replication on of IGF-1 in FXS are also underway.
larger samples. The MAPK/ERK pathway is also Other drugs such as Minocycline (a tetracycline
implicated in FXS pathogenesis. A 16-week open derivative approved for treatment of acne) have been
label trial of Lovastatin in children with FXS sug- shown to have neuroprotective effects and upregu-
gested significant endpoint improvements on clini- late neurotrophic and growth factors in the brain. In
cian-rated measures (Caku, Pellerin, Bouvier, Riou, FXS animal models, Minocycline has been shown to
& Corbin, 2014). However, no placebo-controlled have an inhibitory effect on matrix metallopro-
trials have so far been reported. teinase nine activity which is fundamental in mod-
ulating hippocampal plasticity. A double-blind
mTOR. Mammalian target of rapamycin (mTOR) is placebo controlled cross-over trial of Minocycline
a key regulator in cellular processes including cell in 55 FXS children aged 3.5–16 years over 12 weeks
growth, gene expression and synaptic function. found significant treatment effect on clinician-rated
Clinical Global Impression Scale (CGI) but not on this area, which has already produced some clarifi-
any behavioural measures (Leigh et al., 2013). Fur- cation of neural system pathways and encouraging
thermore, minocycline use was associated with therapeutic leads. These include for instance mTOR
alteration of event-related potentials in an auditory inhibitors, statins, oxytocin and pleiotropic growth
oddball paradigm in 12 subjects taken from the same factors. Despite this, no single treatment has so far
trial (Schneider et al., 2013), possibly suggesting been approved for a Phase III trial and there are
reduced hypersensitivity to auditory stimulation. clearly conceptual as well as practical challenges.
Animal experiments cannot fully recapitulate the
complexity of higher order cognitive processes in
Electrophysiological targets
humans. The interventions target a theoretical
At the more general level of neural system function- underlying pathway towards behavioural outcomes,
ing and long-range integration and connectivity, but it is notable that most studies to date have not
EEG and habituation responses to social stimuli included detailed measures of this hypothesised
have been investigated as targets for ASD interven- pathway and mechanism (Stivaros et al., 2018 is
tion. Linked to underlying social processing and one pilot example that did), and this will be critical in
social attention, these can be some of the earliest the future. Practical challenges include patient
apparent atypicalities in infants at risk who go on to selection, age of treatment onset, dose and duration
develop ASD (Johnson et al., 2015; Szatmari et al., of treatment, dose-limiting side-effects and lack of
2016). As an example of EEG as a target of this kind, CNS biomarkers (Berry-Kravis et al., 2016). Most
Jones, Dawson, Kelly, Estes, & Jane Webb, 2017 human intervention work has been in adolescents or
reported on the EEG outcomes within an RCT of adults and it may be that targeting the developing
parent-delivered psychosocial ‘promoting first rela- brain in much younger samples may be crucial.
tionships’ intervention designed to facilitate parent Outcome measurement in these studies – commonly
infant interaction and delivered between 9– the parent-rated Aberrant Behaviour Checklist
11 months of infant age. Both at 12-month and 18- (ABC; Aman, 1994), or the clinician-rated Global
month follow-ups they reported increase in social Improvement Scale (CGI; Leucht & Engel, 2006),
habituation to faces versus objects and greater Table 2 – have also been very different to that in
increase in frontal EEG theta power (Jones et al., psychosocial research, unfortunately limiting com-
2017). The study was relatively small with valid data parability. This is important because to be convinc-
available from 15 or less in the PFR group and 10 or ing, treatment studies in this area will need to aspire
less in the control group over the range of assess- to the same outcome standards as discussed above
ments, but the findings provide proof of principle for psychosocial research. The need for a common
and suggestive results that neurophysiological mea- outcomes framework is further discussed below.
sures can be impacted by relatively low intensity
social communication focused intervention. The
results echo a previous study (Dawson et al., Summary
2012), which studied EEG in 60% of the trial cases Our focus on target mechanisms for intervention
at endpoint only after two years intensive ESDM across various pathways to autism outcome aims to
therapy. Although the findings showed no interven- give an integrated vantage-point to consider the
tion effect across a number of parameters, there was breadth of current autism intervention research and
reported increase EEG-theta to social stimuli with the continuum of targets across social and biolog-
faces, which correlated with a parent-reported social ical development. In that context, the targets of the
communication rating (although not with autism psychosocial approaches to intervention that we
symptoms, IQ, language or adaptive behaviour). have reviewed can be seen as focusing on endpoint
pathways nearest in development to autism symp-
tom outcomes. They target child-environment social
Summary of neural mechanism targets
interaction and child behaviour within developmen-
Studies of syndromic autism have provided a way of tal processes. We have seen from many studies that
identifying causal links from genes to specific molec- targeted psychosocial interventions of various kinds
ular and biological pathways and alterations in can positively improve over the short-term the most
behavioural outcomes. An underlying hypothesis is immediate parent-child social communication in
that there may be a finite number of common development; but the review may have surprised in
molecular pathways in autism that could act as showing how limited at present is the further
intervention targets and that treatments thus iden- evidence of generalised treatment effect. While we
tified in the context of syndromic ASD could then be do not need more studies to replicate these imme-
applicable to broader idiopathic ASD; thus raising diate effects on their own, there are therefore
the possibility of a precision medicine approach to pressing research questions as to how and whether
autism based on the underlying pathophysiology such proximal communication change can be gen-
(Loth, Murphy, & Spooren, 2016). The work reviewed eralised to improve child social functioning beyond
above reflects the active international effort now in the dyad (generalisation across context and person)
Table 2 Summary of key Biological Mechanism targets and Intervention Outcomes (see text)
Autism Primary
Target group outcome measures Intervention effects
Glutamate receptors
Fenobam (mGluR antagonist) FXS Prepulse inhibition Reduction in anxiety and
Single-dose, and continuous hyperactivity
single-arm, open label study, n = 12 performance test
(Berry-Kravis et al., 2009) to measure
sensory gating
and attention
Mavoglurant (non-competitive mGluR FXS ABC No effect on primary or
inhibitor). Phase 2b multi-centre RCT, secondary outcome measures
n = 175 adults & 139
adolescents stratified
by methylation status
(Berry-Kravis et al., 2016)
Memantine (NMDA receptor antagonist) FXS CGI Improvements on CGI but no
Open-label study, n = 6 improvements on symptom
(Erickson et al., 2009) specific rating scales
GABA
Arbaclofen (GABA agonist) FXS Social avoidance No effect on primary outcome
(Berry-Kravis et al., 2017). n = 125 subscale measure. In child study,
adolescents/adults and 172 children. of FXS specific ABC significant effect in
Phase 3 RCT treatment group noted in
irritability subscale of ABC
and parenting stress index.
Acamprosate (GABA agonist & mGluR FXS CGI Some improvement in social
antagonist) (Erickson et al., 2010), n = 12 behaviour, inattention and
Open-label study hyperactivity
Bumetanide (enhances Non-syndromic CARS Significant improvement in
GABAergic inhibition) autism the treatment group on CARS
(Lemonnier et al., 2017), n = 88.
Multi-centre phase 2 RCT
Dopamine
Methylphenidate FXS Parent & teacher Improvement noted in
(Dopamine re-uptake inhibitor) behaviour attention, social skills and
(Hagerman et al., 1988), n = 15, checklists hyperactivity on
double blind cross-over design methylphenidate
Aripiprazole (partial D2 agonist) FXS CGI Improvement in irritability
(Erickson, Early et al., 2011;
Erickson, Stigler et al., 2011),
n = 12, open label study
Serotonin
Selective Serotonin Re-uptake Non-syndromic CGI and obsessive- No overall positive effect
inhibitors (SSRI’s) autism compulsive behaviours of SSRI’s
(Cochrane review Williams
et al., 2013); nine RCTs of
SSRI’s, n = 320
Oxytocin
Intranasal oxytocin Non-syndromic SRS Significant improvement in
(Parker et al., 2017), n = 32, ASD treatment group when
double blind RCT stratified by baseline blood
oxytocin levels
Signalling pathways
Simvastatin (down NF1 CBCL, WISC No effect on the primary
regulates Ras/MAPK outcome measures
pathway activity)
(van der Vaart et al., 2013),
n = 84, double blind RCT
12 weeks simvastatin NF1-autism Peripheral Simvastatin effects in brain
versus placebo in young pMAPK, MRS, areas previously associated
children (Stivaros et al., 2018), n = 30, ASL, ADC, rsfMRI, with NF1 pathophysiology
4.5–9.5 years, triple blind RCT ABC, CGI and the social brain network
Lovastatin (down regulates NF1 Paired associate No effect
Ras/MAPK pathway learning task
activity) (Payne et al., 2016), n = 146,
double blind RCT
(continued)
Table 2 (continued)
Autism Primary
Target group outcome measures Intervention effects
Everolimus (mTOR inhibitor) TSC Trial evaluated reduction in Reduction in seizure frequency
(Krueger et al., 2013), seizure frequency, and improvement in ASD
n = 20 open label trial quality of life measures related symptoms
Growth & neurotrophic factors
Insulin-like growth factor 1(IGF-1) Phelan-McDermid ABC Significant improvement in
(Kolevzon et al., 2014), n = 9, double syndrome social impairments and
blind cross-over study design restricted behaviours
associated with treatment
IGF-1 (Khwaja et al., 2014), Rett syndrome Cardiorespiratory Improvement in apnoea,
n = 12, single-arm open label measures, anxiety and mood
EEG and behavioural
outcomes CGI and parental
ratings of symptoms
Minocycline FXS CGI, visual analog scale for Treatment effect noted on
(Leigh et al., 2013), n = 55 rating behavioural symptoms clinician-rated measures
Double blind cross-over trial but not on behavioural
measures
ABC, Aberrant Behaviour Questionnaire; ADC, apparent diffusion; ASL, arterial spin labelling; CARS, Childhood Autism Rating
Scale; CBCL, child behaviour checklist; CGI, Clinical Global Improvement; FXS, Fragile X; MRS, magnetic resonance imaging
spectroscopy; NF1, Neurofibromatosis type 1; pMAPK, phosphorylated MAPKinase assay; rsfMRI, resting state functional MRI; SRS,
Social Responsiveness Scale; TS, Tuberous Sclerosis; WISC, Wechsler intelligence scale for children.
and sustained long-term within development (gen- be less, we still argue that this situation is highly
eralisation over time). Only with these demonstra- unsatisfactory and will hamper progress).
tions can an intervention be truly persuasive as an One question for the future is whether there may
autism treatment. We have seen the recently emerg- be synergy between psychosocial treatments of
ing evidence that this is indeed possible, with demonstrated effect and emerging interventions tar-
generalised symptom effects shown for several years geting biological and neural system targets. No
following psychosocial intervention. But more repli- substantive trial of such synergy has yet been
cation studies of this kind are needed. The evidence published and this remains an aspiration. How
on social communication intervention is now at a might such synergy work? We must imagine the
stage of maturity where we should be looking to developing brain in transaction with the quality of its
focus on the common mechanisms across therapies social experience. A great advantage of reviewing
that mediate treatment effect, and to combine biological and psychosocial treatments together is to
evidenced effective treatment mechanisms into new provide such a perspective. Understanding the com-
implementation approaches combining modular plex transactional pathways to social impairment
aspects of intervention in clinically useful ways and development in this way will work against a
(Marchette & Weisz, 2017). A recent example has reductionism that may follow from linear transla-
been the integration of social communication inter- tional models from animal work. The process of
vention with assisted communication (Kasari, Kai- synergy might be simply additive on treatment
ser, et al., 2014). We do not yet know the limits to outcome. However, there might be multiplicative
developmental improvement that may be possible effects through for instance: (i) timely neural system
from psychosocial intervention in this essentially enhancement allowing enhanced response to psy-
neurodevelopmental disorder. chosocial intervention, for instance the hypotheses
We have then reviewed a new generation of that oxytocin administration might help social learn-
emerging pharmacological treatments that are ing in therapy has been tested in pilot studies; (ii)
addressing underlying ‘upstream’ neural mechanism psychosocial intervention altering neural system
targets associated with ASD. Most of these are pre- targets in a way that could be adjunctive to biological
clinical or early phase trials; but we have suggested intervention or promoting sustained effects on devel-
above that many of the psychosocial trials described opment. Preliminary evidence is provided above of
actually have directly analogous status. (It is a sign psychosocial intervention impacting EEG markers of
of the different standards applied to the two domains social processing at least in the short term. Under-
that, while no one would consider widespread pro- standing the continuities and transactions of mech-
motion of some of the biological treatments on this anism between these psychosocial and biological
basis, the review suggests that some psychosocial targets will help progress, clinical work and a
interventions with similar or lower levels of evidence coherent sense of the task.
are widely implemented. Even if the downside risks Common methods of process and outcome mea-
associated with psychosocial treatments are felt to surement across biological and psychosocial
studies will be important going forward – something intervention science that investigates both mecha-
that has not so far happened; the two fields having nism and effect moderation; a Predictive approach
different traditions of outcome measurement. Some that follows intervention effects downstream and be
argue that autism symptom outcomes are the most shown to change things that are theoretically
appropriate metric since they define the disorder important for developmental science – but also for
and can be measured blind. However, methods of families and individuals; and Pre-emptive in the
symptom measurement have unfortunately to date timeliness of such intervention at a point in devel-
differed between psychosocial and biological inter- opment where it may do most benefit. The fact that
vention domains; psychosocial trials using ADOS this has added urgency now is a testament to the
which addresses autism behavioural symptoms advances that have been made in developing poten-
with high external validity, biological trials using tially effective interventions; a success that neces-
ABC or CARS which have less specificity for diag- sitates wide dialogue between basic science,
nostic symptoms and external validity. Others argue interventionists, ethicists, families and the wider
that more ‘real world’ adaptive assessments are autism community, and which also holds real
appropriate, although there is a lack of blinded promise for the developmental outcomes of children
assessments of this kind. There are no appropriate with an autism development.
assessments as yet of child and parent quality of life
in autism, which could help look at wider outcome
(Jonsson et al., 2017; McConachie et al., 2015). Acknowledgements
Choice of assessment relates also to an emerging The authors have declared that they have no competing
debate on the appropriate ultimate aims for inter- or potential conflicts of interest.
vention in autism; an interesting issue that extends
beyond the solely professional or scientific arena.
Applying the ‘4 P’s’ approach to autism intervention Correspondence
(Insel, 2007, 2014) will involve a Participatory Jonathan Green, Child and Adolescent Psychiatry,
approach whereby experts-by-experience of autism University of Manchester, Room 3.311, Jean McFarlane
Building, Oxford Road, Manchester M13 9PL, UK;
should join with interventionists in thinking about
Email: jonathan.green@manchester.ac.uk
the most beneficial intervention outcomes; a Per-
sonalised approach that will be facilitated by quality
Key points
The overall quality of trials evidence in autism psychosocial intervention remains relatively low, despite some
recent progress. Many or even most treatments in current common use have little or no rigorous evidence
base. This is very concerning in such an important disorder.
A variety of psychosocial interventions can demonstrate some short-term effects on children’s immediate
dyadic social interactions, for instance with caregivers. But showing effectiveness in this developmental
disorder requires generalisation of such effects into wider social contexts, on objective autism symptoms or
adaptation and in long-term progress in development. Only a few interventions so far have begun to test or
show this.
A number of early phase interventions on biological targets have shown real promise, but none has yet
progressed to larger scale effectiveness trials on behavioural or symptom outcomes.
There has been enough progress in psychosocial treatment research now that evidence-based practice in
autism can begin to be identified. To consolidate and improve outcomes, the next phase of intervention
research needs improved trial design, and an iterative approach building on success. It could include the
testing of potential synergies between promising biological and psychosocial interventions.
mutations in the SLC29A4 plasma membrane monoamine

References transporter (PMAT) gene. Molecular Autism, 5, 43.
Acosta, M., Kardel, P., Walsh, K., Rosenbaum, K., Gioia, G., & Aldred, C., Green, J., & Adams, C. (2004). A new social
Packer, R. (2011). Lovastatin as treatment for neurocogni- communication intervention for children with autism: Pilot
tive deficits in Neurofibromatosis Type 1: Phase 1 study. randomised controlled treatment study suggesting effective-
Pediatric Neurology, 45, 241–245. ness. Journal of Child Psychology and Psychiatry, 45, 1420–
Adamsen, D., Ramaekers, V., Ho, H.T., Britschgi, C., Rufe- 1430.
nacht, V., Meili, D., . . . & Thony, B. (2014). Autism spectrum Aldred, C., Green, J., Emsley, R., & McConachie, H. (2012).
disorder associated with low serotonin in CSF and Brief report: Mediation of treatment effect in a
communication intervention for pre-school children with Significance of abnormalities in developmental trajectory
autism. Journal of Autism and Developmental Disorders, 42, and asymmetry of cortical serotonin synthesis in autism.
447–454. International Journal of Developmental Neuroscience, 23,
Aman, M. (1994). Aberrant Behaviour Checklist- Community. 171–182.
East Aurora, NY: Slosson Educational Publications. Chang, Y.C., Shire, S.Y., Shih, W., Gelfand, C., & Kasari, C.
Baranek, G.T., Watson, L.R., Turner-Brown, L., Field, S.H., (2016). Preschool deployment of evidence-based social com-
Crais, E.R., Wakeford, L., . . . & Reznick, J.S. (2015). munication intervention: JASPER in the classroom. Journal
Preliminary efficacy of adapted responsive teaching for of Autism and Developmental Disorders, 46, 2211–2223.
infants at risk of autism spectrum disorder in a community Constantino, J. (2002). The social responsiveness scale. Los
sample. Autism Research and Treatment, 2015, 386951. Angeles: Western Psychological Services.
Bear, M.F., Huber, K.M., & Warren, S.T. (2004). The mGluR Cross, S., Kim, S.J., Weiss, L.A., Delahanty, R.J., Sutcliffe,
theory of fragile X mental retardation. Trends in Neuro- J.S., Leventhal, B.L., . . . & Veenstra-Vanderweele, J. (2008).
sciences, 27, 370–377. Molecular genetics of the platelet serotonin system in first-
Bearden, C.E., Hellemann, G.S., Rosser, T., Montojo, C., degree relatives of patients with autism. Neuropsychophar-
Jonas, R., Enrique, N., . . . & Silva, A.J. (2016). A randomized macology, 33, 353–360.
placebo-controlled lovastatin trial for neurobehavioral func- Dadds, M.R., MacDonald, E., Cauchi, A., Williams, K., Levy, F.,
tion in neurofibromatosis I. Annals of Clinical and Transla- & Brennan, J. (2014). Nasal oxytocin for social deficits in
tional Neurology, 3, 266–279. childhood autism: A randomized controlled trial. Journal of
Berry-Kravis, E., Des Portes, V., Hagerman, R., Jacquemont, Autism and Developmental Disorders, 44, 521–531.
S., Charles, P., Visootsak, J., . . . & von Raison, F. (2016). Davies, D.M., de Vries, P.J., Johnson, S.R., McCartney, D.L.,
Mavoglurant in fragile X syndrome: Results of two random- Cox, J.A., Serra, A.L., . . . & Sampson, J.R. (2011). Sirolimus
ized, double-blind, placebo-controlled trials. Science Trans- therapy for angiomyolipoma in tuberous sclerosis and
lational Medicine, 8, 321ra325. sporadic lymphangioleiomyomatosis: A phase 2 trial. Clini-
Berry-Kravis, E., Hagerman, R., Visootsak, J., Budimirovic, D., cal Cancer Research, 17, 4071–4081.
Kaufmann, W.E., Cherubini, M., . . . & Carpenter, R.L. Dawson, G., Jones, E.J., Merkle, K., Venema, K., Lowy, R.,
(2017). Arbaclofen in fragile X syndrome: Results of phase Faja, S., . . . & Webb, S.J. (2012). Early behavioral interven-
3 trials. Journal of Neurodevelopmental Disorders, 9, 3. tion is associated with normalized brain activity in young
Berry-Kravis, E., Hessl, D., Coffey, S., Hervey, C., Schneider, children with autism. Journal of the American Academy of
A., Yuhas, J., . . . & Hagerman, R. (2009). A pilot open label, Child and Adolescent Psychiatry, 51, 1150–1159.
single dose trial of fenobam in adults with fragile X Dawson, G., Rogers, S., Munson, J., Smith, M., Winter, J.,
syndrome. Journal of Medical Genetics, 46, 266–271. Greenson, J., . . . & Varley, J. (2010). Randomized, controlled
Berry-Kravis, E., Hessl, D., Rathmell, B., Zarevics, P., Cheru- trial of an intervention for toddlers with autism: The Early
bini, M., Walton-Bowen, K., . . . & Hagerman, R.J. (2012). Start Denver Model. Pediatrics, 125, e17–e23.
Effects of STX209 (arbaclofen) on neurobehavioral function Domes, G., Heinrichs, M., Glascher, J., Buchel, C., Braus,
in children and adults with fragile X syndrome: A random- D.F., & Herpertz, S.C. (2007). Oxytocin attenuates amygdala
ized, controlled, phase 2 trial. Science Translational Medi- responses to emotional faces regardless of valence. Biological
cine, 4, 152ra127. Psychiatry, 62, 1187–1190.
Biringen, Z., Derscheid, D., Vliegen, N., Closson, L., & Easter- Doussard-Roosevelt, J.A., Joe, C.M., Bazhenova, O.V., &
brooks, A.E. (2014). Emotional availability (EA): Theoretical Porges, S.W. (2003). Mother-child interaction in autistic
background, empirical research using the EA Scales, and and nonautistic children: Characteristics of maternal
clinical applications. Developmental Review, 34, 114–167. approach behaviors and child social responses. Develop-
Blacher, J., Baker, B.L., & Kaladjian, A. (2013). Syndrome ment and Psychopathology, 15, 277–295.
specificity and mother-child interactions: Examining positive Dunn, G., Emsley, R., Liu, H., Landau, S., Green, J., White, I.,
and negative parenting across contexts and time. Journal of & Pickles, A. (2015). Evaluation and validation of social and
Autism and Developmental Disorders, 43, 761–774. psychological markers in randomised trials of complex
Bozdagi, O., Tavassoli, T., & Buxbaum, J.D. (2013). Insulin- interventions in mental health: A methodological research
like growth factor-1 rescues synaptic and motor deficits in a programme. Health Technology Assessment, 19, 51–65.
mouse model of autism and developmental delay. Molecular Erickson, C.A., Early, M., Stigler, K.A., Wink, L.K., Mullett,
Autism, 4, 9. J.E., & McDougle, C.J. (2011). An open-label naturalistic
Brian, J.A., Smith, I.M., Zwaigenbaum, L., & Bryson, S.E. pilot study of acamprosate in youth with autistic disorder.
(2017). Cross-site randomized control trial of the Social ABCs Journal of Child and Adolescent Psychopharmacology, 21,
caregiver-mediated intervention for toddlers with autism 565–569.
spectrum disorder. Autism Research, 10, 1700–1711. Erickson, C.A., Mullett, J.E., & McDougle, C.J. (2009). Open-
Buescher, A.V., Cidav, Z., Knapp, M., & Mandell, D.S. (2014). label memantine in fragile X syndrome. Journal of Autism
Costs of autism spectrum disorders in the United Kingdom and Developmental Disorders, 39, 1629–1635.
and the United States. JAMA Pediatrics, 168, 721–728. Erickson, C.A., Mullett, J.E., & McDougle, C.J. (2010). Brief
Caku, A., Pellerin, D., Bouvier, P., Riou, E., & Corbin, F. report: Acamprosate in fragile X syndrome. Journal of
(2014). Effect of lovastatin on behavior in children and Autism and Developmental Disorders, 40, 1412–1416.
adults with fragile X syndrome: An open-label study. Amer- Erickson, C.A., Stigler, K.A., Wink, L.K., Mullett, J.E., Kohn,
ican Journal of Medical Genetics. Part A, 164A, 2834–2842. A., Posey, D.J., & McDougle, C.J. (2011). A prospective
Carter, A.S., Messinger, D.S., Stone, W.L., Celimli, S., Nah- open-label study of aripiprazole in fragile X syndrome.
mias, A.S., & Yoder, P. (2011). A randomized controlled trial Psychopharmacology (Berl), 216, 85–90.
of Hanen’s ‘More Than Words’ in toddlers with early autism Estes, A., Munson, J., Rogers, S.J., Greenson, J., Winter, J., &
symptoms. Journal of Child Psychology and Psychiatry, 52, Dawson, G. (2015). Long-term outcomes of early interven-
741–752. tion in 6-year-old children with autism spectrum disorder.
Chabernaud, C., Mennes, M., Kardel, P., Gaillard, W., Kalb- Journal of the American Academy of Child and Adolescent
fleisch, L., VanMeter, J., . . . & Acosta, M. (2012). Lovastatin Psychiatry, 54, 580–587.
regulates brain spontaneous low-frequency brain activity in Fletcher-Watson, S., Petrou, A., Scott-Barrett, J., Dicks, P.,
Neurofibromatosis type 1. Neuroscience Letters, 515, 28–33. Graham, C., O’Hare, A., . . . & McConachie, H. (2016). A trial
Chandana, S.R., Behen, M.E., Juhasz, C., Muzik, O., Rother- of an iPad intervention targeting social communication skills
mel, R.D., Mangner, T.J., . . . & Chugani, D.C. (2005). in children with autism. Autism, 20, 771–782.
Freidmann, C., Davis, L., Ciccone, P., & Rubin, R. (1980). Harker, C.M., Ibanez, L.V., Nguyen, T.P., Messinger, D.S., &
Phase 2 double-blind controlled study of a new anxiolytic Stone, W.L. (2016). The effect of parenting style on social
Fenobam vs placebo. Current Therapeutic Research-Clinical smiling in infants at high and low risk for ASD. Journal of
and Experimental, 27, 144–151. Autism and Developmental Disorders, 46, 2399–2407.
French, L., & Kennedy, E.M.M. (2018). Annual Research Hollander, E., Novotny, S., Hanratty, M., Yaffe, R., DeCaria,
Review: Early intervention for infants and young children C.M., Aronowitz, B.R., & Mosovich, S. (2003). Oxytocin
with, or at-risk of, autism spectrum disorder: a systematic infusion reduces repetitive behaviors in adults with autistic
review. Journal of Child Psychology and psychiatry, 59, 444– and Asperger’s disorders. Neuropsychopharmacology, 28,
456. 193–198.
Garg, S., Brooks, A., Burns, A., Burkitt-Wright, E., Kerr, B., Insel, T.R. (2007). The arrival of preemptive psychiatry. Early
Huson, S., . . . & Green, J. (2017). Autism spectrum disorder Intervention in Psychiatry, 1, 5–6.
and other neurobehavioural comorbidities in rare disorders Insel, T.R. (2014). Mental disorders in childhood: Shifting the
of the Ras/MAPK pathway. Developmental Medicine and focus from behavioral symptoms to neurodevelopmental
Child Neurology, 59, 544–549. trajectories. JAMA, 311, 1727–1728.
Garg, S., Green, J., Leadbitter, K., Emsley, R., Lehtonen, A., Johnson, M.H., Gliga, T., Jones, E., & Charman, T. (2015).
Evans, D.G., & Huson, S.M. (2013). Neurofibromatosis type Annual research review: Infant development, autism, and
1 and autism spectrum disorder. Pediatrics, 132, e1642– ADHD–early pathways to emerging disorders. Journal of
e1648. Child Psychology and Psychiatry, 56, 228–247.
Garg, S., Plasschaert, E., Descheemaeker, M.J., Huson, S., Jones, E.J.H., Dawson, G., Kelly, J., Estes, A., & Jane Webb, S.
Borghgraef, M., Vogels, A., . . . & Green, J. (2015). Autism (2017). Parent-delivered early intervention in infants at risk
spectrum disorder profile in neurofibromatosis type I. Jour- for ASD: Effects on electrophysiological and habituation
nal of Autism and Developmental Disorders, 45, 1649–1657. measures of social attention. Autism Research, 10, 961–972.
Green, J. (2015). Editorial: Process to progress? Investigative Jonsson, U., Alaie, I., Lofgren Wilteus, A., Zander, E.,
trials, mechanism and clinical science. Journal of Child Marschik, P.B., Coghill, D., & Bolte, S. (2017). Annual
Psychology and Psychiatry, 56, 1–3. Research Review: Quality of life and childhood mental and
Green, J., Charman, T., McConachie, H., Aldred, C., Slonims, behavioural disorders - a critical review of the research.
V., Howlin, P., . . . & PACT Consortium (2010). Parent- Journal of Child Psychology and Psychiatry, 58, 439–469.
mediated communication-focused treatment in children Juffer, F., Bakermans-Kranenburg, M.J., & van IJzendoorn,
with autism (PACT): A randomised controlled trial. Lancet, M.J. (2008). Promoting Positive Parenting: An Introduction.
375, 2152–2160. In F. Juffer, M.J. Bakermans-Kranenburg, & van IJzendoorn
Green, J., Charman, T., Pickles, A., Wan, M.W., Elsabbagh, M., M.J. (Eds.), Promoting Positive Parenting: An Attachment
Slonims, V., . . . & BASIS Team (2015). Parent-mediated Based Intervention (pp. 1–10). New York: Taylor & Francis.
intervention versus no intervention for infants at high risk of Kaale, A., Fagerland, M.W., Martinsen, E.W., & Smith, L.
autism: A parallel, single-blind, randomised trial. The (2014). Preschool-based social communication treatment for
Lancet. Psychiatry, 2, 133–140. children with autism: 12-month follow-up of a randomized
Green, J., Pickles, A., Pasco, G., Bedford, R., Wan, M.W., trial. Journal of the American Academy of Child and Adoles-
Elsabbagh, M., . . . & British Autism Study of Infant Siblings cent Psychiatry, 53, 188–198.
(BASIS) Team (2017). Randomised trial of a parent-mediated Kaat, A.J., & Farmer, C. (2017). Commentary: Lingering
intervention for infants at high risk for autism: Longitudinal questions about the Social Responsiveness Scale short form.
outcomes to age 3 years. Journal of Child Psychology and A commentary on Sturm et al. (2017). Journal of Child
Psychiatry, 58, 1330–1340. Psychology and Psychiatry, 58, 1062–1064.
Grzadzinski, R., Carr, T., Colombi, C., McGuire, K., Dufek, S., Kasari, C., Gulsrud, A., Paparella, T., Hellemann, G., & Berry,
Pickles, A., & Lord, C. (2016). Measuring changes in social K. (2015). Randomized comparative efficacy study of parent-
communication behaviors: Preliminary development of the mediated interventions for toddlers with autism. Journal of
brief observation of social communication change (BOSCC). Consulting and Clinical Psychology, 83, 554–563.
Journal of Autism and Developmental Disorders, 46, 2464– Kasari, C., Kaiser, A., Goods, K., Nietfeld, J., Mathy, P., Landa,
2479. R., . . . & Almirall, D. (2014). Communication interventions
Guastella, A.J., Gray, K.M., Rinehart, N.J., Alvares, G.A., for minimally verbal children with autism: A sequential
Tonge, B.J., Hickie, I.B., . . . & Einfeld, S.L. (2015). The multiple assignment randomized trial. Journal of the Amer-
effects of a course of intranasal oxytocin on social behaviors ican Academy of Child and Adolescent Psychiatry, 53, 635–
in youth diagnosed with autism spectrum disorders: A 646.
randomized controlled trial. Journal of Child Psychology Kasari, C., Lawton, K., Shih, W., Barker, T.V., Landa, R., Lord,
and Psychiatry, 56, 444–452. C., . . . & Senturk, D. (2014). Caregiver-mediated interven-
Gulsrud, A.C., Hellemann, G., Shire, S., & Kasari, C. (2016). tion for low-resourced preschoolers with autism: An RCT.
Isolating active ingredients in a parent-mediated social Pediatrics, 134, e72–e79.
communication intervention for toddlers with autism spec- Kasari, C., Paparella, T., Freeman, S., & Jahromi, L.B. (2008).
trum disorder. Journal of Child Psychology and Psychiatry, Language outcome in autism: Randomized comparison of
57, 606–613. joint attention and play interventions. Journal of Consulting
Guy, J., Hendrich, B., Holmes, M., Martin, J.E., & Bird, A. and Clinical Psychology, 76, 125–137.
(2001). A mouse Mecp2-null mutation causes neurological Kasari, C., Siller, M., Huynh, L.N., Shih, W., Swanson, M.,
symptoms that mimic Rett syndrome. Nature Genetics, 27, Hellemann, G.S., & Sugar, C.A. (2014). Randomized con-
322–326. trolled trial of parental responsiveness intervention for
Hagerman, R.J., Murphy, M.A., & Wittenberger, M.D. (1988). A toddlers at high risk for autism. Infant Behavior & Develop-
controlled trial of stimulant medication in children with the ment, 37, 711–721.
fragile X syndrome. American Journal of Medical Genetics, Khwaja, O.S., Ho, E., Barnes, K.V., O’Leary, H.M., Pereira,
30, 377–392. L.M., Finkelstein, Y., . . . & Kaufmann, W.E. (2014). Safety,
Hardan, A.Y., Gengoux, G.W., Berquist, K.L., Libove, R.A., pharmacokinetics, and preliminary assessment of efficacy
Ardel, C.M., Phillips, J., . . . & Minjarez, M.B. (2015). A of mecasermin (recombinant human IGF-1) for the treat-
randomized controlled trial of pivotal response treatment ment of Rett syndrome. Proceedings of the National
group for parents of children with autism. Journal of Child Academy of Sciences of the United States of America, 111,
Psychology and Psychiatry, 56, 884–892. 4596–4601.
Knapp, M., Romeo, R., & Beecham, J. (2009). Economic cost of binding in children with autism determined by SPECT.
autism in the UK. Autism, 13, 317–336. Developmental Medicine and Child Neurology, 50, 593–597.
Koegel, L., Singh, A., Koegel, R., Hollingsworth, J., & Brad- Marchette, L.K., & Weisz, J.R. (2017). Practitioner Review:
shaw, J. (2014). Assessing and improving early social Empirical evolution of youth psychotherapy toward transdi-
engagement in infants. Journal of Positive Behavior Inter- agnostic approaches. Journal of Child Psychology and Psy-
ventions, 16, 69–80. chiatry, 58, 970–984.
Kolevzon, A., Angarita, B., Bush, L., Wang, A.T., Frank, Y., McConachie, H., Parr, J.R., Glod, M., Hanratty, J., Livingstone,
Yang, A., . . . & Buxbaum, J.D. (2014). Phelan-McDermid N., Oono, I.P., . . . & Williams, K. (2015). Systematic review of
syndrome: A review of the literature and practice parameters tools to measure outcomes for young children with autism
for medical assessment and monitoring. Journal of Neurode- spectrumdisorder.HealthTechnologyAssessment,19,1–506.
velopmental Disorders, 6, 39. Moessner, R., Marshall, C.R., Sutcliffe, J.S., Skaug, J., Pinto,
Krab, L., de Goede-Bolde, R.A., Aarsen, F., Pluijm, S., Bouman, D., Vincent, J., . . . & Scherer, S.W. (2007). Contribution of
M., van der Geest, J., . . . & Elgersma, Y. (2008). Effect of SHANK3 mutations to autism spectrum disorder. American
simvastatin on cognitive functioning in children with neu- Journal of Human Genetics, 81, 1289–1297.
rofibromatosis type 1: A randomized controlled trial. JAMA, Molosh, A.I., Johnson, P.L., Spence, J.P., Arendt, D., Federici,
300, 287–294. L.M., Bernabe, C., . . . & Shekhar, A. (2014). Social learning
Kraemer, H.C., Wilson, G.T., Fairburn, C.G., & Agras, W.S. and amygdala disruptions in Nf1 mice are rescued by
(2002). Mediators and moderators of treatment effects in blocking p21-activated kinase. Nature Neuroscience, 17,
randomized clinical trials. Archives of General Psychiatry, 1583–1590.
59, 877–883. Mottron, L. (2017). Should we change targets and methods of
Krueger, D.A., Wilfong, A.A., Holland-Bouley, K., Anderson, early intervention in autism, in favor of a strengths-based
A.E., Agricola, K., Tudor, C., . . . & Franz, D.N. (2013). education? European Child and Adolescent Psychiatry, 26,
Everolimus treatment of refractory epilepsy in tuberous 815–825.
sclerosis complex. Annals of Neurology, 74, 679–687. Nakamura, K., Sekine, Y., Ouchi, Y., Tsujii, M., Yoshikawa, E.,
Landa, R.J., Holman, K.C., O’Neill, A.H., & Stuart, E.A. (2011). Futatsubashi, M., . . . & Mori, N. (2010). Brain serotonin and
Intervention targeting development of socially synchronous dopamine transporter bindings in adults with high-func-
engagement in toddlers with autism spectrum disorder: A tioning autism. Archives of General Psychiatry, 67, 59–68.
randomized controlled trial. Journal of Child Psychology and National Institute for Health & Care Excellence. (2013). Autism
Psychiatry, 52, 13–21. spectrum disorder i under 19s: support and management.
Landry, S.H., Smith, K.E., Swank, P.R., & Miller-Loncar, C.L. Available from: https://www.nice.org.uk/guidance/cg170
(2000). Early maternal and child influences on children’s (last accessed 15 December 2017).
later independent cognitive and social functioning. Child Oguro-Ando, A., Rosensweig, C., Herman, E., Nishimura, Y.,
Development, 71, 358–375. Werling, D., Bill, B.R., . . . & Geschwind, D.H. (2015).
Leigh, M.J., Nguyen, D.V., Mu, Y., Winarni, T.I., Schneider, A., Increased CYFIP1 dosage alters cellular and dendritic mor-
Chechi, T., . . . & Hagerman, R.J. (2013). A randomized phology and dysregulates mTOR. Molecular Psychiatry, 20,
double-blind, placebo-controlled trial of minocycline in 1069–1078.
children and adolescents with fragile X syndrome. Journal O’Reilly, C., Lewis, J.D., & Elsabbagh, M. (2017). Is functional
of Developmental and Behavioral Pediatrics, 34, 147–155. brain connectivity atypical in autism? A systematic review of
Lemonnier, E., Villeneuve, N., Sonie, S., Serret, S., Rosier, A., EEG and MEG studies. PLoS ONE, 12, e0175870.
Roue, M., . . . & Ben-Ari, Y. (2017). Effects of bumetanide on Page, M., Wilhelm, M.S., Gamble, W.C., & Card, N.A. (2010). A
neurobehavioral function in children and adolescents with comparison of maternal sensitivity and verbal stimulation as
autism spectrum disorders. Translational Psychiatry, 7, unique predictors of infant social-emotional and cognitive
e1124. development. Infant Behavior & Development, 33, 101–110.
Leucht, S., & Engel, R.R. (2006). The relative sensitivity of the Parker, K.J., Oztan, O., Libove, R.A., Sumiyoshi, R.D., Jack-
Clinical Global Impressions Scale and the Brief Psychiatric son, L.P., Karhson, D.S., . . . & Hardan, A.Y. (2017).
Rating Scale in antipsychotic drug trials. Neuropsychophar- Intranasal oxytocin treatment for social deficits and
macology, 31, 406–412. biomarkers of response in children with autism. Proceedings
Li, W., Cui, Y., Kushner, S., Brown, R., Jentsch, J., Frankland, of the National Academy of Sciences of the United States of
P., . . . & Silva, A. (2005). The HMG-CoA reductase inhibitor America, 114, 8119–8124.
lovastatin reverses the learning and attention deficits in a Parlade, M.V., & Iverson, J.M. (2015). The development of
mouse model of neurofibromatosis type 1. Current Biology, coordinated communication in infants at heightened risk for
15, 1961–1967. autism spectrum disorder. Journal of Autism and Develop-
Lord, C., Risi, S., Lambrecht, L., Cook Jr, E.H., Leventhal, B.L., mental Disorders, 45, 2218–2234.
DiLavore, P.C., . . . & Rutter, M. (2000). The autism diagnos- Payne, J.M., Barton, B., Ullrich, N.J., Cantor, A., Hearps, S.J.,
tic observation schedule-generic: A standard measure of Cutter, G., . . . & NF Clinical Trials Consortium. (2016).
social and communication deficits associated with the Randomized placebo-controlled study of lovastatin in chil-
spectrum of autism. Journal of Autism and Developmental dren with neurofibromatosis type 1. Neurology, 87, 2575–
Disorders, 30, 205–223. 2584.
Loth, E., Murphy, D.G., & Spooren, W. (2016). Defining preci- Pickles, A., Harris, V., Green, J., Aldred, C., McConachie, H.,
sion medicine approaches to autism spectrum disorders: Slonims, V., . . . & PACT Consortium (2015). Treatment
Concepts and challenges. Frontiers in Psychiatry, 7, 188. mechanism in the MRC preschool autism communication
Mahoney, G., Powell, A., & Finger, I. (1986). The maternal trial: Implications for study design and parent-focussed
behavior rating scale. Topics in Early Childhood Special therapy for children. Journal of Child Psychology and
Education, 6, 44–55. Psychiatry, 56, 162–170.
Mainberger, F., Jung, N.H., Zenker, M., Wahllander, U., Pickles, A., Le Couteur, A., Leadbitter, K., Salomone, E., Cole-
Freudenberg, L., Langer, S., . . . & Mall, V. (2013). Lovastatin fletcher, R., Tobin, H., . . . & Green, J. (2016). Parent-
improves impaired synaptic plasticity and phasic alertness mediated social communication therapy for young children
in patients with neurofibromatosis type 1. BMC Neurology, with autism (PACT): A long-term follow-up of a randomised
13, 131. controlled trial. Lancet, 388, 2501–2509.
Makkonen, I., Riikonen, R., Kokki, H., Airaksinen, M.M., & Pinto, D., Delaby, E., Merico, D., Barbosa, M., Merikangas, A.,
Kuikka, J.T. (2008). Serotonin and dopamine transporter Klei, L., & Scherer, S.W. (2014). Convergence of genes and
cellular pathways dysregulated in autism spectrum disor- of dietary and psychological treatments. American Journal of
ders. American Journal of Human Genetics, 94, 677–694. Psychiatry, 170, 275–289.
Poslawsky, I.E., Naber, F.B., Bakermans-Kranenburg, M.J., Stivaros, S., Garg, S., Tziraki, M., Cai, Y., Thomas, O., Mellor,
van Daalen, E., van Engeland, H., & van, I.M.H. (2015). O., . . . & Green, J. (2018). Randomised controlled trial of
Video-feedback Intervention to promote positive parenting simvastatin treatment for autism in young children with
adapted to autism (VIPP-AUTI): A randomized controlled neurofibromatosis type 1 (SANTA). Molecular Autism, 9, 12.
trial. Autism, 19, 588–603. https://doi.org/10.1186/s13229-018-0190-z.
Rahman, A., Divan, G., Hamdani, S.U., Vajaratkar, V., Taylor, Sturm, A., Kuhfeld, M., Kasari, C., & McCracken, J.T. (2017).
C., Leadbitter, K., . . . & Green, J. (2016). Effectiveness of the Development and validation of an item response theory-
parent-mediated intervention for children with autism spec- based Social Responsiveness Scale short form. Journal of
trum disorder in south Asia in India and Pakistan (PASS): A Child Psychology and Psychiatry, 58, 1053–1061.
randomised controlled trial. The Lancet. Psychiatry, 3, 128– Szatmari, P., Chawarska, K., Dawson, G., Georgiades, S.,
136. Landa, R., Lord, C., . . . & Halladay, A. (2016). Prospective
Reichow, B., Barton, E.E., Boyd, B.A., & Hume, K. (2012). longitudinal studies of infant siblings of children with
Early intensive behavioral intervention (EIBI) for young autism: Lessons learned and future directions. Journal of
children with autism spectrum disorders (ASD). Cochrane the American Academy of Child and Adolescent Psychiatry,
Database Systematic Review, (10), CD009260. 55, 179–187.
Rogers, S., Estes, A., Lord, C., Landa, N., Munson, J., & Tomasello, M. (2008). Origins of human communication. Cam-
Dawson, G. (2014). Multisite replication of Early Start bridge, MA: The MIT Press.
Denver Model outcomes. Internation Meeting for Autism Uzunova, G., Hollander, E., & Shepherd, J. (2014). The role of
Research (IMFAR), Abstract 135.003. ionotropic glutamate receptors in childhood neurodevelop-
Rogers, S., Estes, A., Lord, C., Vismara, L., Winter, J., mental disorders: Autism spectrum disorders and fragile X
Fitzpatrick, A., . . . & Dawson, G. (2012). Effects of a brief syndrome. Current Neuropharmacology, 12, 71–98.
Early Start Denver model (ESDM)-based parent intervention van der Vaart, T., Plasschaert, E., Rietman, A.B., Renard, M.,
on toddlers at risk for autism spectrum disorders: A Oostenbrink, R., Vogels, A., . . . & Moll, H.A. (2013). Simvas-
randomized controlled trial. Journal of the American Acad- tatin for cognitive deficits and behavioural problems in
emy of Child and Adolescent Psychiatry, 51, 1052–1065. patients with neurofibromatosis type 1 (NF1-SIMCODA): A
Rutter, M., Bailey, A., & Lord, C. (2003). Social Communication randomised, placebo-controlled trial. The Lancet. Neurology,
Questionnaire. Los Angeles: Western Psychological Services. 12, 1076–1083.
Sameroff, A.J. (2009). The transactional model of development: Veenstra-Vanderweele, J., Jessen, T.N., Thompson, B.J.,
How children and contexts shape each other. Washington, Carter, M., Prasad, H.C., Steiner, J.A., . . . & Blakely, R.D.
DC: American Psychological Association. (2009). Modeling rare gene variation to gain insight into the
Sato, A. (2016). mTOR, a potential target to treat autism oldest biomarker in autism: Construction of the serotonin
spectrum disorder. CNS & Neurological Disorders Drug transporter Gly56Ala knock-in mouse. Journal of Neurode-
Targets, 15, 533–543. velopmental Disorders, 1, 158–171.
Sato, A., Kasai, S., Kobayashi, T., Takamatsu, Y., Hino, O., Wan, M., Brooks, A., Green, J., Abel, K., & Elmadih, A. (2017).
Ikeda, K., & Mizuguchi, M. (2012). Rapamycin reverses Psychometrics and validation of a brief rating measure of
impaired social interaction in mouse models of tuberous parent-infant interaction: Manchester assessment of care-
sclerosis complex. Nature Communications, 3, 1292. giver–infant interaction. International Journal of Behavioral
Schneider, A., Leigh, M.J., Adams, P., Nanakul, R., Chechi, T., Development, 41, 542–549.
Olichney, J., . . . & Hessl, D. (2013). Electrocortical changes Wan, M., Green, J., Elsabbagh, M., Johnson, M., Charman, T.,
associated with minocycline treatment in fragile X syn- & Plummer, F. (2013). Quality of interaction between at-risk
drome. Journal of Psychopharmacology, 27, 956–963. infants and caregiver at 12-15 months is associated with 3-
Shire, S.Y., Chang, Y.C., Shih, W., Bracaglia, S., Kodjoe, M., & year autism outcome. Journal of Child Psychology and
Kasari, C. (2017). Hybrid implementation model of commu- Psychiatry, 54, 763–771.
nity-partnered early intervention for toddlers with autism: A Williams, K., Brignell, A., Randall, M., Silove, N., & Hazell, P.
randomized trial. Journal of Child Psychology and Psychia- (2013). Selective serotonin reuptake inhibitors (SSRIs) for
try, 58, 612–622. autism spectrum disorders (ASD). Cochrane Database Sys-
Siller, M., Hutman, T., & Sigman, M. (2013). A parent- tematic Review, (8), CD004677.
mediated intervention to increase responsive parental Yang, J.C., Rodriguez, A., Royston, A., Niu, Y.Q., Avar, M.,
behaviors and child communication in children with ASD: Brill, R., . . . & Olichney, J.M. (2016). Memantine improves
A randomized clinical trial. Journal of Autism and Develop- attentional processes in fragile X-associated tremor/ataxia
mental Disorders, 43, 540–555. syndrome: Electrophysiological evidence from a randomized
Siller, M., & Sigman, M. (2002). The behaviors of parents of controlled trial. Scientific Reports, 6, 21719.
children with autism predict the subsequent development of Zhou, J., Blundell, J., Ogawa, S., Kwon, C.H., Zhang, W.,
their children’s communication. Journal of Autism and Sinton, C., . . . & Parada, L.F. (2009). Pharmacological
Developmental Disorders, 32, 77–89. inhibition of mTORC1 suppresses anatomical, cellular, and
Sonuga-Barke, E.J., Brandeis, D., Cortese, S., Daley, D., behavioral abnormalities in neural-specific Pten knock-out
Ferrin, M., Holtmann, M., . . . & European, A.G.G. (2013). mice. Journal of Neuroscience, 29, 1773–1783.
Nonpharmacological interventions for ADHD: Systematic
review and meta-analyses of randomized controlled trials Accepted for publication: 18 January 2018

Green Et Al-2018-Journal of Child Psychology and Psychiatry

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Green Et Al-2018-Journal of Child Psychology and Psychiatry

Încărcat de

Drepturi de autor:

Formate disponibile

PFI_12mmX178mm.

pdf + eps format

Journal of Child Psychology and Psychiatry 59:4 (2018), pp 424–443 doi:10.1111/jcpp.12892

Annual Research Review: The state of autism

mechanism level is timely. It points the way forward

in child communication initiation with parent and

Behavioural training methods Summary - proximal dyadic targets

Target Definition and measurement Intervention effects

Parent dyadic behaviours

Post synaptic receptors

mutations in the SLC29A4 plasma membrane monoamine

S-ar putea să vă placă și